If earlier the element base of system power supplies did not raise any questions - they used standard microcircuits, today we are faced with a situation where individual power supply developers begin to produce their own element base, which has no direct analogues among general-purpose elements. One example of this approach is the FSP3528 chip, which is used in a fairly large number of system power supplies manufactured under the FSP trademark.
I had to meet the FSP3528 chip in the following models of system power supplies:
- FSP ATX-300GTF;
- FSP A300F–C;
- FSP ATX-350PNR;
- FSP ATX-300PNR;
- FSP ATX-400PNR;
- FSP ATX-450PNR;
- ComponentPro ATX-300GU.
But since the release of microcircuits makes sense only for mass quantities, you need to be prepared for the fact that it can also be found in other models of FSP power supplies. Direct analogues of this microcircuit have not yet been encountered, therefore, in case of its failure, it is necessary to replace it with exactly the same microcircuit. However, it is not possible to purchase the FSP3528 in a retail network, therefore it can only be found in FSP system power supplies that have been rejected for some other reason.
The FSP3528 chip is available in a 20-pin DIP package (Fig. 1). The purpose of the contacts of the microcircuit is described in Table 1, and Fig. 2 shows its functional diagram. In Table 1, for each output of the microcircuit, the voltage is indicated, which should be on the contact when the microcircuit is typically turned on. A typical application of the FSP3528 chip is its use as part of a submodule for controlling the power supply of a personal computer. This submodule will be discussed in the same article, but a little lower.
|
№ |
Signal |
I/O |
Description |
|
Entrance |
Supply voltage +5V. |
||
|
COMP |
Exit |
Error amplifier output. Inside the microcircuit, the contact is connected to the non-inverting input of the PWM comparator. A voltage is generated at this pin, which is the difference between the input voltages of the error amplifier E/A+ and E/A - (pin 3 and pin 4). During normal operation of the microcircuit, a voltage of about 2.4V is present on the contact. |
|
|
E/A- |
Entrance |
Inverting input of the error amplifier. Inside the microcircuit, this input is biased by 1.25V. The reference voltage of 1.25V is formed by an internal source. During normal operation of the microcircuit, a voltage of 1.23V should be present on the contact. |
|
|
E/A+ |
Entrance |
Non-inverting error amplifier input. This input can be used to control the output voltages of the power supply, i.e. this pin can be considered a feedback signal input. In real circuits, a feedback signal is applied to this pin, obtained by summing all the output voltages of the power supply (+3.3 V /+5V /+12V ). During normal operation of the microcircuit, a voltage of 1.24V should be present on the contact. |
|
|
TREM |
Signal delay control pin ON / OFF (power supply control signal). A time-setting capacitor is connected to this pin. If the capacitor has a capacitance of 0.1 uF, then the turn-on delay ( tone ) is about 8 ms (during this time, the capacitor is charged to a level of 1.8V), and the turn-off delay ( Toff ) is about 24 ms (during this time, the voltage on the capacitor during its discharge decreases to 0.6V). During normal operation of the microcircuit, a voltage of about + 5V should be present on this contact. |
||
|
Entrance |
Power supply on/off signal input. Specification for power supply connectors ATX this signal is referred to as PS-ON. R.E.M. signal is a signal TTL and compared by an internal comparator with a reference level of 1.4V. If the signal REM falls below 1.4V, the PWM chip starts up and the power supply starts working. If the signal REM is set to a high level (more than 1.4V), then the microcircuit is turned off, and, accordingly, the power supply is turned off. On this pin, the voltage can reach a maximum value of 5.25V, although the typical value is 4.6V. During operation, a voltage of about 0.2V should be observed on this contact. |
||
|
Frequency setting resistor of the internal oscillator. During operation, there is a voltage on the contact, about 1.25V. |
|||
|
Frequency setting capacitor of the internal oscillator. During operation, a sawtooth voltage should be observed on the contact. |
|||
|
Entrance |
Overvoltage detector input. The signal from this pin is compared by an internal comparator with an internal reference voltage. This input can be used to control the supply voltage of the microcircuit, to control its reference voltage, as well as to organize any other protection. In typical use, this pin should have a voltage of approximately 2.5V during normal operation of the chip. |
||
|
Signal conditioning delay control pin PG (Power Good) ). A timing capacitor is connected to this pin. A 2.2 µF capacitor provides a 250 ms time delay. The reference voltages for this timing capacitor are 1.8V (when charging) and 0.6V (when discharging). Those. when the power supply is turned on, the signal PG is set to a high level at the moment when the voltage on this timing capacitor reaches a value of 1.8V. And when the power supply is turned off, the signal PG is set to a low level at the moment when the capacitor is discharged to a level of 0.6V. The typical voltage at this pin is +5V. |
|||
|
Exit |
Power good signal - nutrition is normal. A high signal level means that all output voltages of the power supply correspond to the nominal values, and the power supply is operating normally. A low signal level means a power supply failure. The state of this signal during normal operation of the power supply is + 5V. |
||
|
VREF |
Exit |
High-precision reference voltage with a maximum tolerance of ±2%. The typical value of this reference voltage is 3.5 V. |
|
|
V 3.3 |
Entrance |
Overvoltage protection signal in the +3.3 V channel. Voltage is supplied directly to the input from the +3.3 channel V. |
|
|
Entrance |
Overvoltage protection signal in the +5 V channel. Voltage is supplied directly to the input from the +5 channel V. |
||
|
V 12 |
Entrance |
Overvoltage protection signal in the +12 V channel. Voltage is supplied to the input from the +12 channel V through a resistive divider. As a result of using a divider, a voltage of approximately 4.2V is set on this contact (provided that in channel 12 V voltage is +12.5V) |
|
|
Entrance |
Input for additional overvoltage protection signal. This input can be used to organize protection on any other voltage channel. In practical circuits, this contact is used, most often, for short circuit protection in channels -5 V and -12 V . In practical circuits, a voltage of about 0.35V is set on this contact. When the voltage rises to 1.25V, the protection is activated and the microcircuit is blocked. |
||
|
"Earth" |
|||
|
Entrance |
Input for adjusting the "dead" time (the time when the output pulses of the microcircuit are inactive - see Fig. 3). The non-inverting input of the internal dead time comparator is internally biased by 0.12 V. This allows you to set the minimum value of the "dead" time for the output pulses. The "dead" time of the output pulses is regulated by applying to the input DTC DC voltage ranging from 0 to 3.3V. The higher the voltage, the shorter the duty cycle and the longer the dead time. This contact is often used to form a "soft" start when the power supply is turned on. In practical circuits, a voltage of approximately 0.18V is set on this pin. |
||
|
Exit |
The collector of the second output transistor. After starting the microcircuit, pulses are formed on this contact, which follow in antiphase to the pulses on contact C1. |
||
|
Exit |
The collector of the first output transistor. After starting the microcircuit, pulses are formed on this contact, which follow in antiphase to the pulses on contact C2. |
Fig.3 Main parameters of impulses
The FSP3528 chip is a PWM controller designed specifically to control a push-pull pulse converter of a personal computer system power supply. The features of this chip are:
- the presence of built-in protection against overvoltage in the channels + 3.3V / + 5V / + 12V;
- the presence of built-in protection against overload (short circuit) in the channels + 3.3V / + 5V / + 12V;
- the presence of a multi-purpose entrance for the organization of any protection;
- support for the function of switching on the power supply by the input signal PS_ON;
- the presence of a built-in circuit with hysteresis for generating a PowerGood signal (power is normal);
- the presence of a built-in precision source of reference voltages with a tolerance of 2%.
In those power supply models that were listed at the very beginning of the article, the FSP3528 chip is located on the power supply control submodule board. This submodule is located on the secondary side of the power supply and is a printed circuit board placed vertically, i.e. perpendicular to the main board of the power supply (Fig. 4).
This submodule contains not only the FSP3528 chip, but also some elements of its "strapping" that ensure the operation of the chip (see Fig. 5).
The control submodule board is double sided. On the back side of the board there are surface mount elements - SMD, which, by the way, give the most problems due to the not very high quality of soldering. The submodule has 17 contacts arranged in one row. The purpose of these contacts is presented in Table 2.
|
№ |
Contact assignment |
|
Output rectangular pulses designed to control the power transistors of the power supply |
|
|
Power Supply Start Input ( PS_ON ) |
|
|
Channel voltage control input +3.3 V |
|
|
Channel voltage control input +5 V |
|
|
Channel voltage control input +12 V |
|
|
Short circuit protection input |
|
|
Not used |
|
|
Power Good signal output |
|
|
Voltage regulator cathode AZ431 |
|
|
AZ 431 |
|
|
Regulator Reference Input AZ 431 |
|
|
Voltage regulator cathode AZ431 |
|
|
Earth |
|
|
Not used |
|
|
Supply voltage VCC |
On the control submodule board, in addition to the FSP3528 chip, there are two more controlled stabilizers AZ431(analogue of TL431) which are in no way connected with the FSP3528 PWM controller itself, and are designed to control circuits located on the main power supply board.
As an example of the practical implementation of the FSP3528 chip, Fig. 6 shows a diagram of the FSP3528-20D-17P submodule. This control submodule is used in FSP ATX-400PNF power supplies. It is worth noting that instead of a diode D5, a jumper is installed on the board. This sometimes confuses individual specialists who are trying to install a diode in the circuit. Installing a diode instead of a jumper does not change the performance of the circuit - it must function both with a diode and without a diode. However, the installation of the diode D5 can reduce the sensitivity of the short circuit protection circuit.
Such submodules are, in fact, the only example of the use of the FSP3528 chip, so the failure of the submodule elements is often mistaken for a malfunction of the microcircuit itself. In addition, it often happens that specialists fail to identify the cause of the malfunction, as a result of which a microcircuit malfunction is assumed, and the power supply is put aside in the “far corner” or even written off.
In fact, the failure of the microcircuit is a rather rare phenomenon. Elements of the submodule, and, first of all, semiconductor elements (diodes and transistors) are much more likely to fail.
To date, the main malfunctions of the submodule can be considered:
- failure of transistors Q1 and Q2;
- failure of the capacitor C1, which may be accompanied by its "swelling";
- failure of diodes D3 and D4 (simultaneously or separately).
Failure of the remaining elements is unlikely, however, in any case, if you suspect a submodule malfunction, you must first check the soldering of the SMD components on the PCB side of the board.
The diagnostics of the FSP3528 controller is no different from the diagnostics of all other modern PWM controllers for system power supplies, which we have repeatedly talked about on the pages of our magazine. But still, once again, in general terms, we will tell you how you can make sure that the submodule is working.
To check, it is necessary to disconnect the power supply unit with the diagnosed submodule from the mains, and apply all the necessary voltages to its outputs ( +5V, +3.3V, +12V, -5V, -12V, +5V_SB). This can be done using jumpers from another, serviceable, system power supply. Depending on the power supply circuit, it may also be necessary to supply a separate supply voltage +5V on pin 1 of the submodule. This can be done using a jumper between pin 1 of the submodule and the line +5V.
At the same time, on contact CT(cont. 8) a sawtooth voltage should appear, and on the contact VREF(terminal 12) a constant voltage should appear +3.5V.
Next, you need to close the signal to ground PS-ON. This is done by shorting to ground either the contact of the output connector of the power supply (usually a green wire), or pin 3 of the submodule itself. At the same time, rectangular pulses should appear at the output of the submodule (pin 1 and pin 2) and at the output of the FSP3528 chip (pin 19 and pin 20), following in antiphase.
The absence of pulses indicates a malfunction of the submodule or microcircuit.
I would like to note that when using such diagnostic methods, it is necessary to carefully analyze the circuitry of the power supply, since the verification method may change somewhat, depending on the configuration of the feedback circuits and protection circuits from emergency operation of the power supply.
If earlier the element base of system power supplies did not raise any questions - they used standard microcircuits, today we are faced with a situation where individual power supply developers begin to produce their own element base, which has no direct analogues among general-purpose elements. One example of this approach is the FSP3528 chip, which is used in a fairly large number of system power supplies manufactured under the FSP trademark.
I had to meet the FSP3528 chip in the following models of system power supplies:
- FSP ATX-300GTF;
- FSP A300F–C;
- FSP ATX-350PNR;
- FSP ATX-300PNR;
- FSP ATX-400PNR;
- FSP ATX-450PNR;
- ComponentPro ATX-300GU.
But since the release of microcircuits makes sense only for mass quantities, you need to be prepared for the fact that it can also be found in other models of FSP power supplies. Direct analogues of this microcircuit have not yet been encountered, therefore, in case of its failure, it is necessary to replace it with exactly the same microcircuit. However, it is not possible to purchase the FSP3528 in a retail network, therefore it can only be found in FSP system power supplies that have been rejected for some other reason.
The FSP3528 chip is available in a 20-pin DIP package (Fig. 1). The purpose of the contacts of the microcircuit is described in Table 1, and Fig. 2 shows its functional diagram. In Table 1, for each output of the microcircuit, the voltage is indicated, which should be on the contact when the microcircuit is typically turned on. A typical application of the FSP3528 chip is its use as part of a submodule for controlling the power supply of a personal computer. This submodule will be discussed in the same article, but a little lower.
|
№ |
Signal |
I/O |
Description |
|
Entrance |
Supply voltage +5V. |
||
|
COMP |
Exit |
Error amplifier output. Inside the microcircuit, the contact is connected to the non-inverting input of the PWM comparator. A voltage is generated at this pin, which is the difference between the input voltages of the error amplifier E/A+ and E/A - (pin 3 and pin 4). During normal operation of the microcircuit, a voltage of about 2.4V is present on the contact. |
|
|
E/A- |
Entrance |
Inverting input of the error amplifier. Inside the microcircuit, this input is biased by 1.25V. The reference voltage of 1.25V is formed by an internal source. During normal operation of the microcircuit, a voltage of 1.23V should be present on the contact. |
|
|
E/A+ |
Entrance |
Non-inverting error amplifier input. This input can be used to control the output voltages of the power supply, i.e. this pin can be considered a feedback signal input. In real circuits, a feedback signal is applied to this pin, obtained by summing all the output voltages of the power supply (+3.3 V /+5V /+12V ). During normal operation of the microcircuit, a voltage of 1.24V should be present on the contact. |
|
|
TREM |
Signal delay control pin ON / OFF (power supply control signal). A time-setting capacitor is connected to this pin. If the capacitor has a capacitance of 0.1 uF, then the turn-on delay ( tone ) is about 8 ms (during this time, the capacitor is charged to a level of 1.8V), and the turn-off delay ( Toff ) is about 24 ms (during this time, the voltage on the capacitor during its discharge decreases to 0.6V). During normal operation of the microcircuit, a voltage of about + 5V should be present on this contact. |
||
|
Entrance |
Power supply on/off signal input. Specification for power supply connectors ATX this signal is referred to as PS-ON. R.E.M. signal is a signal TTL and compared by an internal comparator with a reference level of 1.4V. If the signal REM falls below 1.4V, the PWM chip starts up and the power supply starts working. If the signal REM is set to a high level (more than 1.4V), then the microcircuit is turned off, and, accordingly, the power supply is turned off. On this pin, the voltage can reach a maximum value of 5.25V, although the typical value is 4.6V. During operation, a voltage of about 0.2V should be observed on this contact. |
||
|
Frequency setting resistor of the internal oscillator. During operation, there is a voltage on the contact, about 1.25V. |
|||
|
Frequency setting capacitor of the internal oscillator. During operation, a sawtooth voltage should be observed on the contact. |
|||
|
Entrance |
Overvoltage detector input. The signal from this pin is compared by an internal comparator with an internal reference voltage. This input can be used to control the supply voltage of the microcircuit, to control its reference voltage, as well as to organize any other protection. In typical use, this pin should have a voltage of approximately 2.5V during normal operation of the chip. |
||
|
Signal conditioning delay control pin PG (Power Good) ). A timing capacitor is connected to this pin. A 2.2 µF capacitor provides a 250 ms time delay. The reference voltages for this timing capacitor are 1.8V (when charging) and 0.6V (when discharging). Those. when the power supply is turned on, the signal PG is set to a high level at the moment when the voltage on this timing capacitor reaches a value of 1.8V. And when the power supply is turned off, the signal PG is set to a low level at the moment when the capacitor is discharged to a level of 0.6V. The typical voltage at this pin is +5V. |
|||
|
Exit |
Power good signal - nutrition is normal. A high signal level means that all output voltages of the power supply correspond to the nominal values, and the power supply is operating normally. A low signal level means a power supply failure. The state of this signal during normal operation of the power supply is + 5V. |
||
|
VREF |
Exit |
High-precision reference voltage with a maximum tolerance of ±2%. The typical value of this reference voltage is 3.5 V. |
|
|
V 3.3 |
Entrance |
Overvoltage protection signal in the +3.3 V channel. Voltage is supplied directly to the input from the +3.3 channel V. |
|
|
Entrance |
Overvoltage protection signal in the +5 V channel. Voltage is supplied directly to the input from the +5 channel V. |
||
|
V 12 |
Entrance |
Overvoltage protection signal in the +12 V channel. Voltage is supplied to the input from the +12 channel V through a resistive divider. As a result of using a divider, a voltage of approximately 4.2V is set on this contact (provided that in channel 12 V voltage is +12.5V) |
|
|
Entrance |
Input for additional overvoltage protection signal. This input can be used to organize protection on any other voltage channel. In practical circuits, this contact is used, most often, for short circuit protection in channels -5 V and -12 V . In practical circuits, a voltage of about 0.35V is set on this contact. When the voltage rises to 1.25V, the protection is activated and the microcircuit is blocked. |
||
|
"Earth" |
|||
|
Entrance |
Input for adjusting the "dead" time (the time when the output pulses of the microcircuit are inactive - see Fig. 3). The non-inverting input of the internal dead time comparator is internally biased by 0.12 V. This allows you to set the minimum value of the "dead" time for the output pulses. The "dead" time of the output pulses is regulated by applying to the input DTC DC voltage ranging from 0 to 3.3V. The higher the voltage, the shorter the duty cycle and the longer the dead time. This contact is often used to form a "soft" start when the power supply is turned on. In practical circuits, a voltage of approximately 0.18V is set on this pin. |
||
|
Exit |
The collector of the second output transistor. After starting the microcircuit, pulses are formed on this contact, which follow in antiphase to the pulses on contact C1. |
||
|
Exit |
The collector of the first output transistor. After starting the microcircuit, pulses are formed on this contact, which follow in antiphase to the pulses on contact C2. |
Fig.3 Main parameters of impulses
The FSP3528 chip is a PWM controller designed specifically to control a push-pull pulse converter of a personal computer system power supply. The features of this chip are:
- the presence of built-in protection against overvoltage in the channels + 3.3V / + 5V / + 12V;
- the presence of built-in protection against overload (short circuit) in the channels + 3.3V / + 5V / + 12V;
- the presence of a multi-purpose entrance for the organization of any protection;
- support for the function of switching on the power supply by the input signal PS_ON;
- the presence of a built-in circuit with hysteresis for generating a PowerGood signal (power is normal);
- the presence of a built-in precision source of reference voltages with a tolerance of 2%.
In those power supply models that were listed at the very beginning of the article, the FSP3528 chip is located on the power supply control submodule board. This submodule is located on the secondary side of the power supply and is a printed circuit board placed vertically, i.e. perpendicular to the main board of the power supply (Fig. 4).
This submodule contains not only the FSP3528 chip, but also some elements of its "strapping" that ensure the operation of the chip (see Fig. 5).
The control submodule board is double sided. On the back side of the board there are surface mount elements - SMD, which, by the way, give the most problems due to the not very high quality of soldering. The submodule has 17 contacts arranged in one row. The purpose of these contacts is presented in Table 2.
|
№ |
Contact assignment |
|
Output rectangular pulses designed to control the power transistors of the power supply |
|
|
Power Supply Start Input ( PS_ON ) |
|
|
Channel voltage control input +3.3 V |
|
|
Channel voltage control input +5 V |
|
|
Channel voltage control input +12 V |
|
|
Short circuit protection input |
|
|
Not used |
|
|
Power Good signal output |
|
|
Voltage regulator cathode AZ431 |
|
|
AZ 431 |
|
|
Regulator Reference Input AZ 431 |
|
|
Voltage regulator cathode AZ431 |
|
|
Earth |
|
|
Not used |
|
|
Supply voltage VCC |
On the control submodule board, in addition to the FSP3528 chip, there are two more controlled stabilizers AZ431(analogue of TL431) which are in no way connected with the FSP3528 PWM controller itself, and are designed to control circuits located on the main power supply board.
As an example of the practical implementation of the FSP3528 chip, Fig. 6 shows a diagram of the FSP3528-20D-17P submodule. This control submodule is used in FSP ATX-400PNF power supplies. It is worth noting that instead of a diode D5, a jumper is installed on the board. This sometimes confuses individual specialists who are trying to install a diode in the circuit. Installing a diode instead of a jumper does not change the performance of the circuit - it must function both with a diode and without a diode. However, the installation of the diode D5 can reduce the sensitivity of the short circuit protection circuit.
Such submodules are, in fact, the only example of the use of the FSP3528 chip, so the failure of the submodule elements is often mistaken for a malfunction of the microcircuit itself. In addition, it often happens that specialists fail to identify the cause of the malfunction, as a result of which a microcircuit malfunction is assumed, and the power supply is put aside in the “far corner” or even written off.
In fact, the failure of the microcircuit is a rather rare phenomenon. Elements of the submodule, and, first of all, semiconductor elements (diodes and transistors) are much more likely to fail.
To date, the main malfunctions of the submodule can be considered:
- failure of transistors Q1 and Q2;
- failure of the capacitor C1, which may be accompanied by its "swelling";
- failure of diodes D3 and D4 (simultaneously or separately).
Failure of the remaining elements is unlikely, however, in any case, if you suspect a submodule malfunction, you must first check the soldering of the SMD components on the PCB side of the board.
The diagnostics of the FSP3528 controller is no different from the diagnostics of all other modern PWM controllers for system power supplies, which we have repeatedly talked about on the pages of our magazine. But still, once again, in general terms, we will tell you how you can make sure that the submodule is working.
To check, it is necessary to disconnect the power supply unit with the diagnosed submodule from the mains, and apply all the necessary voltages to its outputs ( +5V, +3.3V, +12V, -5V, -12V, +5V_SB). This can be done using jumpers from another, serviceable, system power supply. Depending on the power supply circuit, it may also be necessary to supply a separate supply voltage +5V on pin 1 of the submodule. This can be done using a jumper between pin 1 of the submodule and the line +5V.
At the same time, on contact CT(cont. 8) a sawtooth voltage should appear, and on the contact VREF(terminal 12) a constant voltage should appear +3.5V.
Next, you need to close the signal to ground PS-ON. This is done by shorting to ground either the contact of the output connector of the power supply (usually a green wire), or pin 3 of the submodule itself. At the same time, rectangular pulses should appear at the output of the submodule (pin 1 and pin 2) and at the output of the FSP3528 chip (pin 19 and pin 20), following in antiphase.
The absence of pulses indicates a malfunction of the submodule or microcircuit.
I would like to note that when using such diagnostic methods, it is necessary to carefully analyze the circuitry of the power supply, since the verification method may change somewhat, depending on the configuration of the feedback circuits and protection circuits from emergency operation of the power supply.
Assembled
texvedkom.org
A computer power supply, along with such advantages as small dimensions and weight with a power of 250 W or more, has one significant drawback - shutdown in case of overcurrent. This drawback does not allow using the PSU as a charger for a car battery, since the latter has a charging current of several tens of amperes at the initial time. Adding a current limiting circuit to the PSU will avoid turning it off even in the event of a short circuit in the load circuits.
The car battery is charged at constant voltage. With this method, the voltage of the charger remains constant during the entire charging time. Charging the battery in this way is in some cases preferable, since it provides a faster bringing the battery to a state that allows the engine to be started. The energy reported at the initial stage of the charge is spent mainly on the main charging process, that is, on the restoration of the active mass of the electrodes. The strength of the charging current at the initial moment can reach 1.5C, however, for serviceable, but discharged car batteries, such currents will not bring harmful consequences, and the most common ATX PSUs with a power of 300-350 W are not able to deliver a current of more than 16-20A without consequences for themselves. .
The maximum (initial) charging current depends on the model of the PSU used, the minimum limiting current is 0.5A. The idle voltage is adjustable and can be 14 ... 14.5V to charge the starter battery.
First, it is necessary to modify the PSU itself by disabling its protection for exceeding voltages of + 3.3V, + 5V, + 12V, -12V, as well as removing components not used for the charger.
For the manufacture of the memory, the PSU of the FSP ATX-300PAF model was selected. The scheme of the secondary circuits of the PSU was drawn according to the board, and despite a thorough check, minor errors, unfortunately, are not ruled out.
The figure below shows a diagram of an already modified PSU.
For convenient work with the PSU board, the latter is removed from the case, all the wires of the power supply circuits + 3.3V, + 5V, + 12V, -12V, GND, + 5Vsb, feedback wire + 3.3Vs, signal circuit PG, circuit turn on the PSON PSU, power the fan + 12V. Instead of a passive power factor correction choke (installed on the PSU cover), a jumper is temporarily soldered, the ~ 220V power wires coming from the switch on the back of the PSU are soldered out of the board, the voltage will be supplied by the power cord.
First of all, we deactivate the PSON circuit to turn on the PSU immediately after the mains voltage is applied. To do this, instead of the elements R49, C28, we install jumpers. We remove all the elements of the key that supplies power to the T2 galvanic isolation transformer that controls the power transistors Q1, Q2 (not shown in the diagram), namely R41, R51, R58, R60, Q6, Q7, D16. On the power supply board, the contact pads of the collector and emitter of transistor Q6 are connected by a jumper.
After that, we supply ~ 220V to the PSU, make sure it is turned on and works normally.
Next, turn off the control of the -12V power circuit. We remove elements R22, R23, C50, D12 from the board. Diode D12 is located under the group stabilization inductor L1, and it is impossible to remove it without dismantling the latter (it will be written about the alteration of the inductor below), but this is not necessary.
We remove the elements R69, R70, C27 of the PG signal circuit.
Then the overvoltage protection + 5V is disabled. To do this, pin 14 of the FSP3528 (terminal pad R69) is connected by a jumper to the + 5Vsb circuit.
A conductor is cut out on the printed circuit board, connecting pin 14 with the + 5V circuit (elements L2, C18, R20).
The elements L2, C17, C18, R20 are soldered.
We turn on the PSU, make sure it works.
We turn off the protection for overvoltage + 3.3V. To do this, we cut out a conductor on the printed circuit board connecting pin 13 of the FSP3528 with the + 3.3V circuit (R29, R33, C24, L5).
We remove elements of the rectifier and magnetic stabilizer L9, L6, L5, BD2, D15, D25, U5, Q5, R27, R31, R28, R29, R33, VR2, C22, C25, C23, C24, as well as elements of the OOS circuit from the PSU board R35, R77, C26. After that, we add a divider from 910 Ohm and 1.8 kOhm resistors, which forms a voltage of 3.3V from the + 5Vsb source. The middle point of the divider is connected to pin 13 of the FSP3528, the output of the 931 Ohm resistor (a 910 Ohm resistor is suitable) is connected to the + 5Vsb circuit, and the output of the 1.8 kOhm resistor is connected to ground (pin 17 FSP3528).
Further, without checking the operability of the PSU, we turn off the protection along the + 12V circuit. Unsolder the chip resistor R12. In the contact pad R12, connected to the pin. 15 FSP3528 a 0.8 mm hole is drilled. Instead of resistor R12, a resistance is added, consisting of series-connected resistors with a nominal value of 100 ohms and 1.8 kOhm. One resistance output is connected to the + 5Vsb circuit, the other to the R67 circuit, pin. 15 FSP3528.
We solder the elements of the OOS circuit + 5V R36, C47.
After removing the OOS in the + 3.3V and + 5V circuits, it is necessary to recalculate the value of the OOS resistor in the + 12V R34 circuit. The reference voltage of the error amplifier FSP3528 is 1.25V, with the variable resistor VR1 in the middle position, its resistance is 250 ohms. With a voltage at the PSU output of +14V, we get: R34 = (Uout / Uop - 1) * (VR1 + R40) = 17.85 kOhm, where Uout, V is the output voltage of the PSU, Uop, V is the reference voltage of the FSP3528 error amplifier (1.25V), VR1 is the resistance of the tuning resistor, Ohm, R40 is the resistance of the resistor, Ohm. The value of R34 is rounded up to 18 kOhm. Set up for a fee.
It is advisable to replace the C13 3300x16V capacitor with a 3300x25V capacitor and add the same one to the place freed from C24 in order to divide the ripple currents between them. The positive output of C24 is connected to the + 12V1 circuit through a choke (or jumper), the + 14V voltage is removed from the + 3.3V contact pads.
We turn on the PSU, by adjusting VR1 we set the output voltage to + 14V.
After all the changes made to the BP, we move on to the limiter. The current limiter circuit is shown below.
Resistors R1, R2, R4 ... R6 connected in parallel form a current-measuring shunt with a resistance of 0.01 Ohm. The current flowing in the load causes a voltage drop on it, which the DA1.1 op-amp compares with the reference voltage set by the tuning resistor R8. A DA2 stabilizer with an output voltage of 1.25V is used as a reference voltage source. Resistor R10 limits the maximum voltage applied to the error amplifier to 150 mV, which means the maximum load current to 15A. The limiting current can be calculated by the formula I \u003d Ur / 0.01, where Ur, V is the voltage on the R8 engine, 0.01 Ohm is the shunt resistance. The current limiting circuit works as follows.
The output of the error amplifier DA1.1 is connected to the output of the resistor R40 on the power supply board. As long as the allowable load current is less than that set by resistor R8, the voltage at the output of the op-amp DA1.1 is zero. The PSU is operating normally, and its output voltage is determined by the expression: Uout = ((R34/(VR1+R40))+1)*Uop. However, as the voltage on the measuring shunt increases due to an increase in the load current, the voltage at pin 3 of DA1.1 tends to the voltage at pin 2, which leads to an increase in the voltage at the output of the op-amp. The PSU output voltage begins to be determined by another expression: Uout=((R34/(VR1+R40))+1)*(Uop-Uosh), where Uosh, V is the voltage at the output of the error amplifier DA1.1. In other words, the PSU output voltage begins to decrease until the current flowing in the load becomes slightly less than the set limiting current. The state of equilibrium (current limitation) can be written as follows: Ush/Rsh=(((R34/(VR1+R40))+1)*(Uop-Ush))/Rn, where Rsh, Ohm – shunt resistance, Ush, V – shunt drop voltage, Rн, Ohm – load resistance.
Op-amp DA1.2 is used as a comparator, signaling with the help of the HL1 LED to turn on the current limiting mode.
The printed circuit board (under the "iron") and the layout of the elements of the current limiter are shown in the figures below.
A few words about the details and their replacement. It makes sense to replace the electrolytic capacitors installed on the FSP power supply board with new ones. First of all, in the rectifier circuits of the standby power supply + 5Vsb, these are C41 2200x10V and C45 1000x10V. Do not forget about boosting capacitors in the base circuits of power transistors Q1 and Q2 - 2.2x50V (not shown in the diagram). If possible, it is better to replace the 220V (560x200V) rectifier capacitors with new, larger ones. Capacitors of the 3300x25V output rectifier must be of low ESR - WL or WG series, otherwise they will quickly fail. In extreme cases, you can put used capacitors of these series for a lower voltage - 16V.
Precision op amp DA1 AD823AN "rail-to-rail" fits this circuit perfectly. However, it can be replaced by an order of magnitude cheaper op-amp LM358N. At the same time, the stability of the output voltage of the PSU will be somewhat worse, you will also have to select the value of the resistor R34 down, since this op-amp has a minimum output voltage instead of zero (0.04V, to be precise) 0.65V.
The maximum total power dissipation of the current measuring resistors R1, R2, R4…R6 KNP-100 is 10 W. In practice, it is better to limit yourself to 5 watts - even at 50% of the maximum power, their heating exceeds 100 degrees.
Diode assemblies BD4, BD5 U20C20, if they really cost 2 pieces, it makes no sense to change to something more powerful, they hold well as promised by the manufacturer of PSU 16A. But it happens that in reality only one is installed, in which case it is necessary either to limit the maximum current to 7A, or to add a second assembly.
Testing the PSU with a current of 14A showed that after 3 minutes the temperature of the L1 inductor winding exceeds 100 degrees. Long-term trouble-free operation in this mode raises serious doubts. Therefore, if it is intended to load the PSU with a current of more than 6-7A, it is better to redo the inductor.
In the factory version, the +12V choke winding is wound with a single-core wire with a diameter of 1.3 mm. The PWM frequency is 42 kHz, with which the depth of current penetration into copper is about 0.33 mm. Due to the skin effect at this frequency, the effective wire cross-section is no longer 1.32 mm2, but only 1 mm2, which is not enough for a current of 16A. In other words, a simple increase in the diameter of the wire to obtain a larger cross section, and therefore reduce the current density in the conductor, is inefficient for this frequency range. For example, for a wire with a diameter of 2 mm, the effective cross section at a frequency of 40 kHz is only 1.73 mm2, and not 3.14 mm2, as expected. For the efficient use of copper, we wind the inductor winding with a litz wire. We will make a litz wire from 11 pieces of enameled wire 1.2 m long and 0.5 mm in diameter. The diameter of the wire may be different, the main thing is that it be less than twice the depth of current penetration into copper - in this case, the wire cross section will be used by 100%. The wires are folded into a “bundle” and twisted with a drill or screwdriver, after which the bundle is threaded into a heat shrink tube with a diameter of 2 mm and crimped with a gas burner.
The finished wire is completely wound around the ring, and the manufactured inductor is installed on the board. It makes no sense to wind the -12V winding, the HL1 “Power” indicator does not require any stabilization.
It remains to install the current limiter board in the PSU case. The easiest way is to screw it to the end of the radiator.
Let's connect the "OOS" circuit of the current regulator to the resistor R40 on the power supply board. To do this, cut out a part of the track on the PSU circuit board, which connects the output of the resistor R40 to the “case”, and next to the contact pad R40 we drill a 0.8mm hole where the wire from the regulator will be inserted.
Let's connect the power supply of the current regulator + 5V, for which we solder the corresponding wire to the + 5Vsb circuit on the PSU board.
The “case” of the current limiter is connected to the “GND” pads on the PSU board, the -14V circuit of the limiter and +14V of the PSU board go to external “crocodiles” for connecting to the battery.
Indicators HL1 "Power" and HL2 "Restriction" are fixed in place of the plug installed instead of the "110V-230V" switch.
Most likely, your outlet does not have a protective earth contact. Or rather, there may be a contact, but the wire does not fit to it. There is nothing to say about the garage ... It is strongly recommended to organize protective grounding at least in the garage (basement, shed). Do not ignore safety precautions. This sometimes ends very badly. For those who do not have a 220V socket, equip the PSU with an external screw terminal to connect it.
After all the improvements, turn on the PSU and adjust the required output voltage with the trimming resistor VR1, and the maximum current in the load with the resistor R8 on the current limiter board.
We connect a 12V fan to the circuits -14V, + 14V of the charger on the power supply board. For normal operation of the fan, two diodes connected in series are switched on in the wire break + 12V or -12V, which will reduce the fan supply voltage by 1.5V.
We connect the passive power factor correction choke, 220V power supply from the switch, screw the board into the case. We fix the output cable of the charger with a nylon tie.
Screw on the lid. The charger is ready to go.
In conclusion, it is worth noting that the current limiter will work with an ATX (or AT) PSU of any manufacturer using PWM controllers TL494, KA7500, KA3511, SG6105 or the like. The difference between them will be only in the methods of bypassing the protections.
Download limiter circuit board in PDF and DWG format (Autocad)
shemopedia.ru
On the Internet, you can find a lot of descriptions and ways to remake the ATX PSU to suit your needs, from chargers to laboratory power supplies. The scheme of the secondary circuits of the ATX PSU from the manufacturer's brand FSP is approximately the same:
It makes no sense to describe the details of the operation of the circuit, since everything is on the network, I will only note that in this circuit there is an adjustment of the short-circuit protection current. - VR3 trimmer, which eliminates the need to add a current detector circuit and a shunt. However, if there is such a need, then you can always add such a section of the circuit, for example, on a simple and common op-amp LM358. Often, in such power supplies as FSP, the PWM controller cascade is made in the form of a module:
As always, the secondary circuits on the board are soldered:
We check the operability of the “duty room” and the serviceability of the power inverter, otherwise pre-repair!
The schematic diagram of a converted 15-35 volt power supply looks like this:
A 47k trimmer resistor sets the required voltage at the feeder output. Highlighted in red on the diagram - delete.
Assembled
The radiator of the rectifier diodes is small in area, so it is better to increase it. According to the results of measurements at a voltage of 28V, the converted PSU easily gave out 7A, given its initial power of 350W, the calculated load voltage:
Of course, it is always possible to buy a ready-made power supply of such power, but given the cost of such devices, a real economic justification for these costs is needed ...
atreds.pw
High quality Chip BA3528FP in our online store at retail and wholesale at a bargain price!
Until recently, the BA3528FP chip offered by our online store was difficult to buy anywhere. But with the advent of specialized stores, such as ours, it has become possible to make purchases in any volume: in a single copy or in a batch with fast delivery throughout Russia!
A flexible payment system will allow you to immediately pay for the order + shipping cost online and save on the transfer of cash on delivery to the account of our store! We will deliver your order by Russian Post or Transport Company to the point of self-delivery or by courier to the door in the shortest possible time.
Save
Read more on Elhow: https://elhow.ru/ucheba/russkij-jazyk/orfografija/pravopisanie-glagolov/sekonomit-ili-sekonomit?utm_source=users&utm_medium=ct&utm_campaign=ct
Previously, our audience was not so large, but today we have expanded our cooperation boundaries and offer products from the best manufacturers for a wide range of customers. And, it does not matter where you live, you can order the BA3528FP chip from any city in our country with the possibility of delivery to any, even its most remote point.
At present, there is fierce competition on the cost - the speed of delivery of orders - we strongly recommend that you choose delivery by a Transport Company. because Its delivery cost, although not significantly higher than that of the Russian Post (about 15-20%), but the speed of work and the absence of queues, as well as a loyal attitude towards the client, is incommensurably higher! :))
There is no doubt about the quality of the offered product Chip BA3528FP from a well-known manufacturer. The BA3528FP meets all high quality standards and is certified at the factory, which is why it is in high demand among many of our customers. One category of consumers uses the BA3528FP chip for personal purposes, others for the purpose of running and developing a business.
For each product, we offer detailed specifications, parameters and instructions for use, so you can choose the lot that suits you and the lot you need Microcircuit BA3528FP model BA3528FP. The presented model takes into account the demand and wishes of customers, takes into account the demand for the product on the market, and constantly updates the product range.
You can find the BA3528FP Chip in the appropriate subcategory - Radio Components/Import Chips/BA using the convenient electronic search. We care about all customers and try to ensure that each customer is satisfied with the product, quality of service, favorable delivery terms, advice, cost. Our plans are to help everyone and everyone, and therefore we offer products only from a trusted manufacturer.
We will carefully pack the BA3528FP chip into your order and deliver it as soon as possible, which is especially important for buyers who need it very urgently. We would like to draw your attention to the fact that the prices for the BA3528FP chip of the BA3528FP model in our online store are the most optimal and affordable. The need for such products arises as needed. You can postpone the purchase of the BA3528FP chip for later, or you can place an order right now, while the price of the goods remains the same - extremely low and profitable. It is always pleasant to make purchases at low prices, especially when the order concerns more than one unit of goods - this allows you to profitably save not only money, but also precious time!
radio-sale.ru
I will continue to publish articles on the technical characteristics of the most popular LEDs. Today, according to the plan, I have to talk about the "old" SMD 3528, or rather about their characteristics. I note that the lighting properties of any diode are constantly improving. Therefore, there may be some discrepancies. Plus, each manufacturer can add something to the detriment of another characteristic. But this is not critical, because. the majority adheres to a single "nomenclature". Each manufacturer has its own Datasheet, but the main characteristics practically do not change.
At the dawn of its appearance, SMD 3528 was widely used in almost all light sources. Starting from indicator devices and ending with lighting lamps. And if they looked more or less tolerable on indicator devices, then LED lamps left much to be desired. There was little light from them (compared to current technologies). I once wrote that 3528 are beginning to become obsolete. Most manufacturers are phasing them out in lighting, automotive, and the like. The process of "leaving" the market is quite lengthy and while these types of diodes can be found in decorative lighting, decorative light bulbs, indicator devices, and of course, there is no way to get away from LED strips. It is precisely due to the tapes used in backlights, due to their tolerable glow and almost no heating, that SMD 3528 continues to "cling" to the rapidly developing LED market.
The LED is produced with one crystal. As a result, we get one color: either all shades of white, or colored diodes - red, green, blue, yellow.
The lens used in production is transparent. The chip is based on InGaN. As a rule, the lens consists of a silicone compound. The case is similar in material to SMD 5050.
If we compare the luminous flux with 5050, then for the diodes we are discussing today it is almost three times less and is only 4.5-5 Lumens. Previously, this was a revolutionary value, but now, looking at this data, you want to smile. And smile in a good way. After all, 3528 did their job and gave rise to three-crystal diodes. Therefore, I will not judge them strictly)
I will consider the Datasheet from a Chinese manufacturer, with whom our company constantly works and has no complaints about it yet. At one time they worked only in wholesale lots, but recently they have also gone to retail. Rather small wholesale. The minimum order quantity is 200 pieces. Their price is less than that of Russian sellers, and the quality remains at the same level. We have already produced more than one thousand light sources from the LEDs of this company. And ... well, their delivery is free to Russia. For those who still do not believe that China is quietly producing decent products, it is worth talking to my colleague Konstantin Ogorodnikov, who will tell you "why there are holes in the bread." He shoveled more than one Chinese supplier for us until he found the right ones)
Characteristics of SMD 3528 cool white glow
Characteristic curves of warm white SMD 3528
Since only chips with a white glow are most common, I will omit the Datasheet 3528 SMD with a different color. Yes, it is not necessary. Something tells me that it is unlikely that anyone will be interested in these types of diodes. Well, if suddenly ... Then you will find all the data at the link that you indicated earlier. True, you will have to do the translation yourself. The manufacturer provides the Datasheet in Chinese. But by comparing my pictures with the notation and the Chinese "waste paper" you will easily understand everything and you will be able to create the performance characteristics with your own translation.
Any LED from the SMD series has a four-digit designation. Based on them, we can immediately get information about the size of the chips. the first two are the length, the second are the width. Dimensions are in mm. Different manufacturers have their own errors, but they do not go beyond + -0.1-0.15 mm.
Diodes are produced in 2000 pieces in a cassette (roll). If you are engaged in constant "needlework", then it is more profitable to order by rolls. And more convenient and practical. Especially if you have lamps on these diodes at home and you constantly have to solder them.)
And finally, some cautions when working with any SMD diodes.
This is not my whim or my experience. This is a real warning from the manufacturers!
The vast majority of diodes are coated with a silicone compound. Despite the fact that it is less susceptible to mechanical stress, it must be handled with care:
Well, in principle, and all the simple rules that everyone should follow. And this is where I finish the story about the characteristics of SMD 3528 LEDs and retire to compile another, more interesting material for me. Well, I don’t like to write about obvious things, and even more so, characteristics that every self-respecting person who studied at school should be able to read))).
leds-test.com
If earlier the element base of system power supplies did not raise any questions - they used standard microcircuits, now we are faced with a situation where individual power supply developers begin to produce their own element base, which has no direct analogues among general purpose parts. One example of this approach is the FSP3528 chip, which is used in a fairly large number of system power supplies manufactured under the FSP trademark.
I had to meet the FSP3528 chip in subsequent models of system power supplies:
FSP ATX-300GTF-
FSP A300F–C-
FSP ATX-350PNR-
FSP ATX-300PNR-
FSP ATX-400PNR-
FSP ATX-450PNR-
ComponentPro ATX-300GU.
Fig.1 FSP3528 pinout
But because the release of microcircuits makes sense only for mass quantities, it is necessary to be prepared for the fact that it can also be found in other models of FSP power supplies. Direct analogues of this microcircuit have not yet been encountered, therefore, in the event of its failure, it must be replaced with exactly the same microcircuit. But it is not likely to purchase the FSP3528 in a retail network, therefore it can only be found in FSP system power supplies, rejected by any other judgment.
Fig.2 Multifunctional diagram of the FSP3528 PWM controller
The FSP3528 chip is available in a 20-pin DIP package (Fig. 1). The purpose of the contacts of the microcircuit is described in Table 1, and in Fig. 2 its multifunctional circuit is shown. In Table 1, for each output of the microcircuit, the voltage that should be on the contact for a typical microcircuit is turned on is indicated. And a typical application of the FSP3528 chip is its implementation as part of a computer power supply control submodule. This submodule will be discussed in the same article, but a little lower.
Table 1. Purpose of the contacts of the PWM controller FSP3528
Description
Supply voltage +5V.
Error amplifier output. Inside the microcircuit, the contact is connected to the non-inverting input of the PWM comparator. A voltage is generated at this pin, which is the difference between the input voltages of the error amplifier E / A + and E / A - (pin 3 and pin 4). During normal operation of the microcircuit, a voltage of about 2.4V is present on the contact.
Inverting input of the error amplifier. Inside the microcircuit, this input is shifted by 1.25V. The reference voltage of 1.25V is formed by an internal source. During normal operation of the microcircuit, the contact should have a voltage of 1.23V.
Non-inverting error amplifier input. This input can be used to control the output voltages of the power supply, i.e. this pin can be considered as a feedback signal input. In real circuits, a feedback signal is applied to this pin, obtained by summing all the output voltages of the power supply (+3.3V/+5V/+12V). During normal operation of the microcircuit, the contact should have a voltage of 1.24V.
ON / OFF signal delay control contact (power supply turn-on control signal). A time-setting capacitor is connected to this pin. If the capacitor has a capacitance of 0.1 uF, then the turn-on delay (Ton) is about 8 ms (during this period of time, the capacitor charges to a level of 1.8V), and the turn-off delay (Toff) is about 24 ms (during this period of time, the voltage across the capacitor when it is discharged, it miniaturizes to 0.6V). During normal operation of the microcircuit, this pin should have a voltage of about + 5V.
Power supply on/off signal input. In the specification for ATX power supply connectors, this signal is referred to as PS-ON. The REM signal is a TTL signal and is compared by an internal comparator to a 1.4V reference level. If the REM signal goes below 1.4V, the PWM chip starts up and the power supply starts working. If the REM signal is set to the highest level (more than 1.4V), then the microcircuit is turned off, and the power supply is turned off accordingly. On this pin, the voltage can reach a maximum value of 5.25V, although the typical value is 4.6V. During operation, a voltage of about 0.2V should be observed on this contact.
Frequency setting resistor of the internal oscillator. During operation, there is a voltage on the contact, about 1.25V.
Frequency setting capacitor of the internal oscillator. During operation, a sawtooth voltage should be observed on the contact.
Overvoltage sensor input. The signal from this pin is compared by an internal comparator with an internal reference voltage. This input can be used to control the supply voltage of the microcircuit, to control its reference voltage, and to organize any other protection. In typical use, this pin should have a voltage of approximately 2.5V during normal operation of the chip.
PG signal generation delay control contact (Power Good). A timing capacitor is connected to this pin. A 2.2 µF capacitor provides a 250 ms time delay. The reference voltages for this timing capacitor are 1.8V (when charging) and 0.6V (when discharging). That is, when the power supply is turned on, the PG signal is set to the highest level at the moment when the voltage on this time-setting capacitor reaches 1.8V. And when the power supply is turned off, the PG signal is set to a low level at the moment when the capacitor is discharged to a level of 0.6V. The typical voltage at this pin is +5V.
Power Good signal - power is normal. The highest signal level means that all output voltages of the power supply correspond to the nominal values, and the power supply is operating normally. A low signal level means a malfunction of the power supply. The state of this signal during normal operation of the power supply is + 5V.
High precision reference voltage with a tolerance of less than ±2%. The typical value of this reference voltage is 3.5 V.
Overvoltage protection signal in the +3.3V channel. Voltage is supplied directly to the input from the +3.3V channel.
Overvoltage protection signal in the +5 V channel. Voltage is supplied directly to the input from the + 5V channel.
Overvoltage protection signal in the +12V channel. Voltage is supplied to the input from the +12V channel through a resistive divider. As a result of using a divider, a voltage of approximately 4.2V is set on this contact (provided that the voltage in the 12V channel is + 12.5V)
Input for additional overvoltage protection signal. This input can be used to organize protection on any other voltage channel. In practical circuits, this contact is used, in most cases, to protect against a short circuit in the -5V and -12V channels. In practical circuits, a voltage of about 0.35V is set on this contact. When the voltage rises to 1.25V, the protection is activated and the microcircuit is blocked.
Input for adjusting the "dead" time (the time when the output pulses of the microcircuit are inactive - see Fig. 3). The non-inverting input of the internal dead time comparator is biased by 0.12V internally. This allows you to set a small value of "dead" time for the output pulses. The “dead” time of the output pulses is regulated by applying a constant voltage from 0 to 3.3V to the DTC input. The higher the voltage, the shorter the duty cycle and the longer the dead time. This contact is often used to form a "soft" start when the power supply is turned on. In practical circuits, this pin is set to a voltage of approximately 0.18V.
The collector of the second output transistor. After starting the microcircuit, pulses are formed on this contact, which follow in antiphase to the pulses on contact C1.
The collector of the first output transistor. After starting the microcircuit, pulses are formed on this contact, which follow in antiphase to the pulses on contact C2.
Fig.3 Main characteristics of impulses
The FSP3528 chip is a PWM controller designed specifically to control a push-pull pulse converter of a computer system power supply. The features of this chip are:
The presence of integrated protection against overvoltage in the channels + 3.3V / + 5V / + 12V-
The presence of integrated protection against overload (short circuit) in the channels + 3.3V / + 5V / + 12V-
The presence of a multi-purpose entrance for the organization of any kind of protection -
Support for the function of turning on the power supply by the input signal PS_ON-
The presence of an integrated circuit with hysteresis for generating a PowerGood signal (power is normal) -
The presence of a built-in precision reference voltage source with a tolerance of 2%.
In those power supply models that were listed at the very beginning of the article, the FSP3528 chip is located on the power supply control submodule board. This submodule is located on the secondary side of the power supply and is an integrated circuit placed vertically, i.e. perpendicular to the main board of the power supply (Fig. 4).
Fig.4 Power supply with submodule FSP3528
This submodule contains not only the FSP3528 chip, but also some elements of its “strapping” that ensure the operation of the chip (see Fig. 5).
Fig.5 Submodule FSP3528
The control submodule board has a double-sided installation. On the back side of the board there are surface mount elements - SMD, which, by the way, give the greatest number of problems due to the not very high soldering properties. The submodule has 17 contacts arranged in one row. The purpose of these contacts is presented in Table 2.
Table 2. Purpose of contacts of submodule FSPЗ3528-20D-17P
Purpose of the contact
Output rectangular pulses designed to control the power transistors of the power supply
Power Supply Start Input (PS_ON)
Channel voltage control input +3.3V
Channel voltage control input +5V
Channel voltage control input +12V
Low Fault Input
Not used
Power Good signal output
Reference voltage input of AZ431 regulator
Voltage regulator cathode AZ431
Not used
Supply voltage VCC
On the control submodule board, in addition to the FSP3528 chip, there are two more AZ431 controlled stabilizers (analogue of TL431) that are in no way connected with the FSP3528 PWM controller itself, and are designed to control circuits located on the main power supply board.
As an example of the practical implementation of the FSP3528 chip, Fig. 6 shows a diagram of the FSP3528-20D-17P submodule. This control submodule is used in FSP ATX-400PNF power supplies. It is worth paying attention that instead of the D5 diode, a jumper is installed on the board. This sometimes confuses some professionals who try to install a diode in the circuit. Installing a diode instead of a jumper does not change the operation of the circuit - it should work both with a diode and without a diode. But installing a D5 diode can reduce the sensitivity of the protection circuit against small short circuits.
Fig.6 FSP3528-20D-17P submodule diagram
Such submodules are, in fact, the only example of the implementation of the FSP3528 microcircuit, therefore, the failure of parts of the submodule is often mistaken for a malfunction of the microcircuit itself. In addition, it often happens that the specialists fail to identify the cause of the malfunction, as a result of which the malfunction of the microcircuit is implied, and the power supply is put aside in the “far corner” or is generally written off.
In fact, the failure of the microcircuit is a rather rare phenomenon. The elements of the submodule, and, at first, semiconductor elements (diodes and transistors) are even more often subject to failures.
Today, the main defects of the submodule can be considered:
Failure of transistors Q1 and Q2-
Failure of the capacitor C1, which may be accompanied by its "swelling" -
Failure of diodes D3 and D4 (immediately or separately).
Failure of other parts is unlikely, but in any case, if you suspect a malfunction of the submodule, you must first check the soldering of the SMD components on the PCB side of the board.
Chip diagnostics
Diagnostics of the FSP3528 controller is no different from the diagnostics of all other modern PWM controllers for system power supplies, which we have already known more than once on the pages of our magazine. But still, again, in general terms, we will tell you how you can make sure that the submodule is working.
To check, you need to disconnect the power supply with the diagnosed submodule from the network, and apply all the necessary voltages to its outputs (+5V, +3.3V, +12V, -5V, -12V, +5V_SB). This can be done using jumpers from another, serviceable, system power supply. Depending on the power supply circuit, it may also be necessary to apply a separate +5V supply voltage to pin 1 of the submodule. This can be done using a jumper between pin 1 of the submodule and the + 5V line.
With all this, a sawtooth voltage should appear on the CT pin (pin. 8), and a constant voltage of + 3.5V should appear on the VREF pin (pin. 12).
Next, you need to close the PS-ON signal to the ground. This is done by shorting to ground either the output connector of the power supply (usually a greenish wire), or pin 3 of the submodule itself. With all this, at the output of the submodule (pin 1 and pin 2) and at the output of the FSP3528 chip (pin 19 and pin 20), rectangular pulses should appear, following in antiphase.
The absence of pulses indicates a malfunction of the submodule or microcircuit.
It is worth noting that when using similar diagnostic methods, you need to carefully consider the circuitry of the power supply, because the verification method may change somewhat, depending on the configuration of the feedback circuits and the emergency protection circuits of the power supply.
alunekst.ru
The BA3528AFP/BA3529AFP ICs from ROHM are designed for use in stereo players. They operate at 3V and include a two-channel preamp, a two-channel power amplifier and a motor controller. The on-chip reference voltage source eliminates the need for decoupling capacitors when connecting an audio head and headphones. The motor controller uses a bridge circuit to minimize the number of external components, which improves reliability and reduces the size of the device. Brief electrical characteristics of the BA3528AFP / BA3529AFP microcircuits are shown in Table 1. A typical switching circuit is shown in fig. 1. The input signal from the playback head goes to the non-inverting inputs of the preamplifiers (outputs
Fig.1. Typical wiring diagram m/s BA3528AFP/BA3529AFP
Table 1. Main parameters of m/s BA3528AFP/BA3529AFP
19, 23), and the common wire of the head is connected to a reference voltage source (pin 22). The negative feedback signal is fed from the preamplifier outputs (pins 17, 25) through corrective RC chains to the inverting inputs (pins 19, 24). The amplified signal can be fed to the volume controls through electronic keys (pins 16, 26). The keys are closed if the microcircuit supply voltage is applied to the control input (pin 1). For the BA3529AFP chip, it is possible to enable Dolby noise suppressors in the output circuits of the preamplifiers. After adjusting the level, the audio signal is fed to the output power amplifiers (pins 15, 27) with a fixed gain. Its value is a classification parameter and is 36 dB for BA3528AFP and 27 dB for BA3529AFP. From the outputs of power amplifiers (pins 2, 12), the signal is fed to headphones with a resistance of 16-32 ohms, the common wire of which is connected to a powerful reference voltage source (pin 11). The main factor that reduces the reliability of the microcircuit and leads to its failure is the violation of its power parameters. The manufacturer limits the power dissipated by the microcircuit to 1.7 W at a temperature not exceeding 25 "C, with a decrease in this value by 13.6 mW for each degree of temperature rise. Complete replacements for the BA3528AFP / BA3529AFP microcircuits are the BA3528FP / BA3529FP microcircuits.
nakolene.narod.ru
A computer power supply, along with such advantages as small dimensions and weight with a power of 250 W or more, has one significant drawback - shutdown in case of overcurrent. This drawback does not allow using the PSU as a charger for a car battery, since the latter has a charging current of several tens of amperes at the initial time. Adding a current limiting circuit to the PSU will avoid turning it off even in the event of a short circuit in the load circuits.
The car battery is charged at constant voltage. With this method, the voltage of the charger remains constant during the entire charging time. Charging the battery in this way is in some cases preferable, since it provides a faster bringing the battery to a state that allows the engine to be started. The energy reported at the initial stage of the charge is spent mainly on the main charging process, that is, on the restoration of the active mass of the electrodes. The strength of the charging current at the initial moment can reach 1.5C, however, for serviceable, but discharged car batteries, such currents will not bring harmful consequences, and the most common ATX PSUs with a power of 300-350 W are not able to deliver a current of more than 16-20A without consequences for themselves. .
The maximum (initial) charging current depends on the model of the PSU used, the minimum limiting current is 0.5A. The idle voltage is adjustable and can be 14 ... 14.5V to charge the starter battery.
First, it is necessary to modify the PSU itself by disabling its protection for exceeding voltages of + 3.3V, + 5V, + 12V, -12V, as well as removing components not used for the charger.
For the manufacture of the memory, the PSU of the FSP ATX-300PAF model was selected. The scheme of the secondary circuits of the PSU was drawn according to the board, and despite a thorough check, minor errors, unfortunately, are not ruled out.
The figure below shows a diagram of an already modified PSU.
For convenient work with the PSU board, the latter is removed from the case, all the wires of the power supply circuits + 3.3V, + 5V, + 12V, -12V, GND, + 5Vsb, feedback wire + 3.3Vs, signal circuit PG, circuit turn on the PSON PSU, power the fan + 12V. Instead of a passive power factor correction choke (installed on the PSU cover), a jumper is temporarily soldered, the ~ 220V power wires coming from the switch on the back of the PSU are soldered out of the board, the voltage will be supplied by the power cord.
First of all, we deactivate the PSON circuit to turn on the PSU immediately after the mains voltage is applied. To do this, instead of the elements R49, C28, we install jumpers. We remove all the elements of the key that supplies power to the T2 galvanic isolation transformer that controls the power transistors Q1, Q2 (not shown in the diagram), namely R41, R51, R58, R60, Q6, Q7, D16. On the power supply board, the contact pads of the collector and emitter of transistor Q6 are connected by a jumper.
After that, we supply ~ 220V to the PSU, make sure it is turned on and works normally.
Next, turn off the control of the -12V power circuit. We remove elements R22, R23, C50, D12 from the board. Diode D12 is located under the group stabilization inductor L1, and it is impossible to remove it without dismantling the latter (it will be written about the alteration of the inductor below), but this is not necessary.
We remove the elements R69, R70, C27 of the PG signal circuit.
Then the overvoltage protection + 5V is disabled. To do this, pin 14 of the FSP3528 (terminal pad R69) is connected by a jumper to the + 5Vsb circuit.
A conductor is cut out on the printed circuit board, connecting pin 14 with the + 5V circuit (elements L2, C18, R20).
The elements L2, C17, C18, R20 are soldered.
We turn on the PSU, make sure it works.
We turn off the protection for overvoltage + 3.3V. To do this, we cut out a conductor on the printed circuit board connecting pin 13 of the FSP3528 with the + 3.3V circuit (R29, R33, C24, L5).
We remove elements of the rectifier and magnetic stabilizer L9, L6, L5, BD2, D15, D25, U5, Q5, R27, R31, R28, R29, R33, VR2, C22, C25, C23, C24, as well as elements of the OOS circuit from the PSU board R35, R77, C26. After that, we add a divider from 910 Ohm and 1.8 kOhm resistors, which forms a voltage of 3.3V from the + 5Vsb source. The middle point of the divider is connected to pin 13 of the FSP3528, the output of the 931 Ohm resistor (a 910 Ohm resistor is suitable) is connected to the + 5Vsb circuit, and the output of the 1.8 kOhm resistor is connected to ground (pin 17 FSP3528).
Further, without checking the operability of the PSU, we turn off the protection along the + 12V circuit. Unsolder the chip resistor R12. In the contact pad R12, connected to the pin. 15 FSP3528 a 0.8 mm hole is drilled. Instead of resistor R12, a resistance is added, consisting of series-connected resistors with a nominal value of 100 ohms and 1.8 kOhm. One resistance output is connected to the + 5Vsb circuit, the other to the R67 circuit, pin. 15 FSP3528.
We solder the elements of the OOS circuit + 5V R36, C47.
After removing the OOS in the + 3.3V and + 5V circuits, it is necessary to recalculate the value of the OOS resistor in the + 12V R34 circuit. The reference voltage of the error amplifier FSP3528 is 1.25V, with the variable resistor VR1 in the middle position, its resistance is 250 ohms. With a voltage at the PSU output of +14V, we get: R34 = (Uout / Uop - 1) * (VR1 + R40) = 17.85 kOhm, where Uout, V is the output voltage of the PSU, Uop, V is the reference voltage of the FSP3528 error amplifier (1.25V), VR1 is the resistance of the tuning resistor, Ohm, R40 is the resistance of the resistor, Ohm. The value of R34 is rounded up to 18 kOhm. Set up for a fee.
It is advisable to replace the C13 3300x16V capacitor with a 3300x25V capacitor and add the same one to the place freed from C24 in order to divide the ripple currents between them. The positive output of C24 is connected to the + 12V1 circuit through a choke (or jumper), the + 14V voltage is removed from the + 3.3V contact pads.
We turn on the PSU, by adjusting VR1 we set the output voltage to + 14V.
After all the changes made to the BP, we move on to the limiter. The current limiter circuit is shown below.
Resistors R1, R2, R4 ... R6 connected in parallel form a current-measuring shunt with a resistance of 0.01 Ohm. The current flowing in the load causes a voltage drop on it, which the DA1.1 op-amp compares with the reference voltage set by the tuning resistor R8. A DA2 stabilizer with an output voltage of 1.25V is used as a reference voltage source. Resistor R10 limits the maximum voltage applied to the error amplifier to 150 mV, which means the maximum load current to 15A. The limiting current can be calculated by the formula I \u003d Ur / 0.01, where Ur, V is the voltage on the R8 engine, 0.01 Ohm is the shunt resistance. The current limiting circuit works as follows.
The output of the error amplifier DA1.1 is connected to the output of the resistor R40 on the power supply board. As long as the allowable load current is less than that set by resistor R8, the voltage at the output of the op-amp DA1.1 is zero. The PSU is operating normally, and its output voltage is determined by the expression: Uout = ((R34/(VR1+R40))+1)*Uop. However, as the voltage on the measuring shunt increases due to an increase in the load current, the voltage at pin 3 of DA1.1 tends to the voltage at pin 2, which leads to an increase in the voltage at the output of the op-amp. The PSU output voltage begins to be determined by another expression: Uout=((R34/(VR1+R40))+1)*(Uop-Uosh), where Uosh, V is the voltage at the output of the error amplifier DA1.1. In other words, the PSU output voltage begins to decrease until the current flowing in the load becomes slightly less than the set limiting current. The state of equilibrium (current limitation) can be written as follows: Ush/Rsh=(((R34/(VR1+R40))+1)*(Uop-Ush))/Rn, where Rsh, Ohm – shunt resistance, Ush, V – shunt drop voltage, Rн, Ohm – load resistance.
Op-amp DA1.2 is used as a comparator, signaling with the help of the HL1 LED to turn on the current limiting mode.
Printed circuit board ( under "iron") and the layout of the elements of the current limiter is shown in the figures below.
A few words about the details and their replacement. It makes sense to replace the electrolytic capacitors installed on the FSP power supply board with new ones. First of all, in the rectifier circuits of the standby power supply + 5Vsb, these are C41 2200x10V and C45 1000x10V. Do not forget about boosting capacitors in the base circuits of power transistors Q1 and Q2 - 2.2x50V (not shown in the diagram). If possible, it is better to replace the 220V (560x200V) rectifier capacitors with new, larger ones. Capacitors of the 3300x25V output rectifier must be of low ESR - WL or WG series, otherwise they will quickly fail. In extreme cases, you can put used capacitors of these series for a lower voltage - 16V.
Precision op amp DA1 AD823AN "rail-to-rail" fits this circuit perfectly. However, it can be replaced by an order of magnitude cheaper op-amp LM358N. At the same time, the stability of the output voltage of the PSU will be somewhat worse, you will also have to select the value of the resistor R34 down, since this op-amp has a minimum output voltage instead of zero (0.04V, to be precise) 0.65V.
The maximum total power dissipation of the current measuring resistors R1, R2, R4…R6 KNP-100 is 10 W. In practice, it is better to limit yourself to 5 watts - even at 50% of the maximum power, their heating exceeds 100 degrees.
Diode assemblies BD4, BD5 U20C20, if they really cost 2 pieces, it makes no sense to change to something more powerful, they hold well as promised by the manufacturer of PSU 16A. But it happens that in reality only one is installed, in which case it is necessary either to limit the maximum current to 7A, or to add a second assembly.
Testing the PSU with a current of 14A showed that after 3 minutes the temperature of the L1 inductor winding exceeds 100 degrees. Long-term trouble-free operation in this mode raises serious doubts. Therefore, if it is intended to load the PSU with a current of more than 6-7A, it is better to redo the inductor.
In the factory version, the +12V choke winding is wound with a single-core wire with a diameter of 1.3 mm. The PWM frequency is 42 kHz, with which the depth of current penetration into copper is about 0.33 mm. Due to the skin effect at this frequency, the effective wire cross section is no longer 1.32 mm 2, but only 1 mm 2, which is not enough for a current of 16A. In other words, a simple increase in the diameter of the wire to obtain a larger cross section, and therefore reduce the current density in the conductor, is inefficient for this frequency range. For example, for a wire with a diameter of 2 mm, the effective cross section at a frequency of 40 kHz is only 1.73 mm 2, and not 3.14 mm 2, as expected. For the efficient use of copper, we wind the inductor winding with a litz wire. We will make a litz wire from 11 pieces of enameled wire 1.2 m long and 0.5 mm in diameter. The diameter of the wire may be different, the main thing is that it be less than twice the depth of current penetration into copper - in this case, the wire cross section will be used by 100%. The wires are folded into a “bundle” and twisted with a drill or screwdriver, after which the bundle is threaded into a heat shrink tube with a diameter of 2 mm and crimped with a gas burner.
The finished wire is completely wound around the ring, and the manufactured inductor is installed on the board. It makes no sense to wind the -12V winding, the HL1 “Power” indicator does not require any stabilization.
It remains to install the current limiter board in the PSU case. The easiest way is to screw it to the end of the radiator.
Let's connect the "OOS" circuit of the current regulator to the resistor R40 on the power supply board. To do this, cut out a part of the track on the PSU circuit board, which connects the output of the resistor R40 to the “case”, and next to the contact pad R40 we drill a 0.8mm hole where the wire from the regulator will be inserted.
Let's connect the power supply of the current regulator + 5V, for which we solder the corresponding wire to the + 5Vsb circuit on the PSU board.
The “case” of the current limiter is connected to the “GND” pads on the PSU board, the -14V circuit of the limiter and +14V of the PSU board go to external “crocodiles” for connecting to the battery.
Indicators HL1 "Power" and HL2 "Restriction" are fixed in place of the plug installed instead of the "110V-230V" switch.
Most likely, your outlet does not have a protective earth contact. Or rather, there may be a contact, but the wire does not fit to it. There is nothing to say about the garage ... It is strongly recommended to organize protective grounding at least in the garage (basement, shed). Do not ignore safety precautions. This sometimes ends very badly. For those who do not have a 220V socket, equip the PSU with an external screw terminal to connect it.
After all the improvements, turn on the PSU and adjust the required output voltage with the trimming resistor VR1, and the maximum current in the load with the resistor R8 on the current limiter board.
We connect a 12V fan to the circuits -14V, + 14V of the charger on the power supply board. For normal operation of the fan, two diodes connected in series are switched on in the wire break + 12V or -12V, which will reduce the fan supply voltage by 1.5V.
We connect the passive power factor correction choke, 220V power supply from the switch, screw the board into the case. We fix the output cable of the charger with a nylon tie.
Screw on the lid. The charger is ready to go.
In conclusion, it is worth noting that the current limiter will work with an ATX (or AT) PSU of any manufacturer using PWM controllers TL494, KA7500, KA3511, SG6105 or the like. The difference between them will be only in the methods of bypassing the protections.
Download limiter circuit board in PDF and DWG format (Autocad)
