L&C go further lower with their lying transformers. I've had countless units with ERL-28 mains transformers marked as 35. Example shown below.

Note the outline of the L shape. A true proper 35 size transformer will fill all that empty spot, and to a degree so will a 33. The one shown there is a 28 size transformer and when desoldered, it reads "ERL-28 XP" followed by what is a datecode of when it was produced.

momaka wrote on 2026-05-19, 14:05:

There's generally no penalty for going with higher capacitance (even if doubling it), but sometimes there is for going too low with the ESR (certain older PSUs like this may oscillate very badly with caps with too low ESR.) Therefore, DO NOT use polymers.

It depends on where in the circuit it is.
The polymers work perfectly well with the +5V standby power supply.
Apparently, the frequency response of the standby power supply is so slow that it usually doesn’t require any correction at all.
For the powering PWM IC standby supply, I generally use a stack of 10 x 1μF 50V MLCC X7R capacitors.
Solder them in a vertical column 2*5 and fit a heat-shrink sleeve.
I just happened to have a bag of 100 of them in a drawer in my desk. 😀

momaka wrote on 2026-04-27, 22:21:

If it was mine, I'd recap and save this one, especially if it's an APFC-less unit (if anyone hasn't gotten the hint already, I have a strong distaste for APFC). It's probably not that bad of a unit. Bulged Fuhjyyus are just an ugly sight, but nothing that can't be remedied. Make sure to recap the 5VSB circuit completely, though, as I think this one is old enough to have the 2-transistor self-oscillating design. They go bad in these FSP PSUs when the "critical" small cap on the primary for that circuit fails. The result is destroyed 5VSB section.

But once recapped, should be fine for a retro 5V-heavy PC.
FWIW, I remove the "dummy" load resistors on these on the 5V rail, because they produce too much heat and cook the output caps unnecessarily.

momaka wrote on 2026-05-01, 17:02:

You can replace just the big caps on the output / secondary side, as CharlieFoxTrot suggested, if that will make it easier for you. The smaller caps on the daughterboards indeed very rarely fail, so can usually be left alone. I only replace them on PSUs I really care about and/or any PSU that has a circuit(s) that may critically depends on such small caps (e.g. some 2-transistor 5VSB circuits, among others.)

So I started recapping this but in the middle of it I noticed that I had forgotten to buy 4.7µF caps. So is the exact value of that cap important here? Ie. do I need to put the old Teapo cap back, or can I use say, a 10µF cap instead? It appears to be in middle of a resistor divider.

Mike_ wrote on 2026-05-21, 18:55:

So I started recapping this but in the middle of it I noticed that I had forgotten to buy 4.7µF caps. So is the exact value of that cap important here? Ie. do I need to put the old Teapo cap back, or can I use say, a 10µF cap instead? It appears to be in middle of a resistor divider.

It looks like the anode of that cap goes to a resistor which then goes to a potentiometer.
If I remember correctly, one of the pots was for adjusting the voltage level, while the other was for adjust the OPP level. However, I'm not sure which one this cap appears to be in circuit with. So to be on the safer side, it's probably better that you put back the original 4.7 uF Teapo cap... unless it is badly failed, of course.
-OR-
If you have spare 10 uF caps in stock, you can take two 10 uF caps and connect them in series like batteries ('-' on first cap to ground, '+' on first cap to '-' on 2nd cap, and '+' on 2nd cap to '+' on board.) For two 10 uF caps in series, you will get 5 uF, which is more than reasonably close.

PcBytes wrote on 2026-05-19, 23:44:

L&C go further lower with their lying transformers. I've had countless units with ERL-28 mains transformers marked as 35.

Yeah, I think I have one or two PSUs like that as well, and one was probably an L&C indeed. Though other (cheap) manufacturers did it too.

That said, the size of the transformer rarely worries me that much. It's just that it's an indication for the rest of the PSU will likely also be built with really cheap/undersized parts. Apart from that, though, you can still make a decent PSU even with an ERL-28 transformer. Sure, the power rating of the PSU will be much more limited (generally around 150W-180W max.) But honestly, even that should be enough for a single socket Pentium II/3 and lower-power Athlon systems.

The output filtering on the other hand, is one place that will matter the most.
- If it's a PSU that has only one output cap per rail and no free spots to add more (the missing) caps to those rails, then you can forget about turning such PSU into anything useful for powering a PC. (Now, if you want to power some DC motors with such PSU, by all means not a single one will care about the bad ripple 😉 ).
- If the PSU is missing the rod inductors between the caps (the "PI coils") and there is no space on the PCB to add them, then that will also further degrade the filtering (well, in most cases anyways - some PSUs and certain rails don't always have or need these... but that will depend on the design.)

So all in all, as long as the output filtering is OK, the rest of the PSU isn't that interesting/concerning... unless, of course, everything is so undersized that the PSU can't even do 100W of power.

shevalier wrote on 2026-05-20, 03:32:

It depends on where in the circuit it is.
The polymers work perfectly well with the +5V standby power supply.

Good call about the 5VSB circuit!
That is indeed just about always the exception when it comes to recapping, where any low to ultra-low ESR cap will do.
I also regularly do the 5VSB with ultra-low ESR caps - not because of any benefits, but because I still have a lingering (and aging) supply of 16V 1500 uF caps from various scrapped Xbox 360 motherboards. Most 5VSB circuits (but especially the 2-transistor self-oscillating ones) are usually more than happy with those. The ones with TNY or TOP switching ICs sometimes become a little louder in operation (the 5VSB will squeal / produce a high-pitched sound). They still work and regulate fine, of course.

shevalier wrote on 2026-05-20, 03:32:

For the powering PWM IC standby supply, I generally use a stack of 10 x 1μF 50V MLCC X7R capacitors.
Solder them in a vertical column 2*5 and fit a heat-shrink sleeve.
I just happened to have a bag of 100 of them in a drawer in my desk. 😀

Nice!
... but too "rich" for my taste... or rather, I'd probably find something else to do with those 50V 10 uF caps.
Though what I have done before is fit once such ceramic cap (25V 4.7 uF) in parallel with the existing 25V 47 uF "critical" cap on the 5VSB circuit of a 250W Deer PSU, just to see what would happen. Short answer: nothing unusual - the 5VSB worked fine.

shevalier wrote on 2026-05-19, 16:21:

Because in the original Delta (or DTK? Whoever made the first classic half-bridge AT, which everyone then started copying), protection was based on the base current of the transistors rather than a current transformer in the primary circuit of the main transformer.

Yes.
On the "better" half-bridge PSUs, there's a 4th small transformer that exists purely to relay primary-side current sensing to the secondary so that OPP works more accurately.
I see that done very rarely, though.

shevalier wrote on 2026-05-19, 16:21:

I may have got the number of turns spot on, but the voltages were within 3.3 ±0.15 V.
Which is terrible for SDRAM and the corresponding chipsets.
For newer systems, well, that’s problem of the voltage regulators on the motherboard. 😀

True.
Indeed only motherboards that don't generate their own 3.3V supply for the RAM can be adversely affected by this.
For most even remotely newer boards (typically starting with Pentium II era), most chips will have their own dedicated supply on the board, so the voltage of the 3.3V rail isn't so relevant anymore.

momaka wrote on 2026-05-21, 21:58:

... but too "rich" for my taste...

If the power capacitors in the secondary circuits dry out, the power supply unit will usually simply stop working properly.
If the standby capacitors dry out, the entire power supply unit usually burns out. Often, the motherboard goes with it.
And taking a power supply unit apart is hardly a pleasant experience.
I prefer to do it once and for all.

True.
Indeed only motherboards that don't generate their own 3.3V supply for the RAM can be adversely affected by this.
For most even remotely newer boards (typically starting with Pentium II era), most chips will have their own dedicated supply on the board, so the voltage of the 3.3V rail isn't so relevant anymore.

I have never come across a single S370-based motherboard on which the +3.3 V power supply has been implemented properly.
- Either a regulated step-up DC/DC converter supplying 3.3 V to the entire board,
- or a separate regulator for the memory/chipset, with all slots powered directly from the PSU.

Usually, it’s some sort of odd design.
- Asus-style: DC/DC converters with a +5 V output; the PSU’s 3.3 V output isn’t used at all
- Gigabyte style: separate linear regulators for memory and expansion slots.

As a result, the power supply goes haywire from such an uneven load.
That’s why I prefer boards that use the +3.3 V bus directly.

momaka wrote on 2026-05-21, 21:58:

It looks like the anode of that cap goes to a resistor which then goes to a potentiometer. If I remember correctly, one of the p […]
Show full quote

It looks like the anode of that cap goes to a resistor which then goes to a potentiometer.
If I remember correctly, one of the pots was for adjusting the voltage level, while the other was for adjust the OPP level. However, I'm not sure which one this cap appears to be in circuit with. So to be on the safer side, it's probably better that you put back the original 4.7 uF Teapo cap... unless it is badly failed, of course.
-OR-
If you have spare 10 uF caps in stock, you can take two 10 uF caps and connect them in series like batteries ('-' on first cap to ground, '+' on first cap to '-' on 2nd cap, and '+' on 2nd cap to '+' on board.) For two 10 uF caps in series, you will get 5 uF, which is more than reasonably close.

I put two 2.2µF caps in parallel, I guess that should work too even if it doesn't look pretty. 😁

Anyways, I tested the PSU a bit and it seems to work fine now.

Now and then I 'stay on the lookout' for the Zalman 400 Watt models, ZM400B-APS.
40A on the +5V rail and a P4 + AUX connector. I find them quite reliable, heavy build and silent.

momaka wrote on 2026-05-19, 14:05:

Therefore, DO NOT use polymers.

From personal experience.

The FSP ATX-350PAF power supplies work perfectly with Capson LZ 1800 μF × 16 V × 13 mΩ capacitors (electrolytic capacitors with ultra-low ESR). These capacitors are not of particularly high quality in themselves, but their ESR is comparable to the figures in the datasheet, at least when it comes to new and genuine units.
There were no other 10mm diameter capacitors available. 🙁

This general-purpose polymer capacitor from Kemet has, in principle, very similar technical specifications. (10mOhm)
https://eu.mouser.com/datasheet/3/72/1/A750MW … 28M1CAAE010.pdf
I don't have capacitors of that value on hand, so I haven't specifically tested them.

The experiments were carried out on a Hipro HP-D5201AW PSU 520 W (APFC, 2 transistors forward , Champion CM6800TX PWM controller, group regulation).
I tried general use polymer caps 1500 µF 10V on the +5V rail and 2200 µF 6.3V caps on the +3.3V rail, and they work perfectly.
The 820 µF*16V capacitors work fine, but their capacitance is critically insufficient.
The power supply stability issues started precisely when I added five more capacitors to the two 820 µF capacitors to achieve the standard capacitance of 2200 µF*2.
Polymer capacitors have a very weak dependence of their ESR on capacitance. This means you can't replace 2200 µF (12 mOhm) with 3 x 680 µF, because 680 µF have an ESR of 13 or 14 mOhm, not 36 mOhm.

So, most likely, if you use the same number of polymer cap as electrolytic cap and the same capacity, there shouldn't be any problems.
But experimentation is necessary.

So that Codegen did not work from the get go. Didn't turn on properly.
Turns out 4 caps where dead. One for the negative 12V rail, 2 for the standby PSU (primary and secondary) and one on the 12V rail. Luckily I had all I needed in Stock and now it runs well. I'll replace the remaining caps on the 5V and 3.3V rails tomorrow for good measure.

shevalier wrote on 2026-04-23, 08:35:

Isn't this PSU actually an EOL? And then there’s the cooling. For some people, with three 120mm fans in the case, it’ll manage e […]
Show full quote

Mike_ wrote on 2026-04-23, 07:55:

It's not quite *that* bad. For aforementioned RM550x 3,3V/5V outputs have combined total power of 130W and from RM750x up it's 150W. It's not optimal, but should be usable for most old ATX setups.

https://assets.corsair.com/image/upload/corsa … 2018_Manual.pdf

Isn't this PSU actually an EOL?
And then there’s the cooling.
For some people, with three 120mm fans in the case, it’ll manage even with an ASUS TUSL2-C or CUBX.
But for others, even with a Duron, PSU go into thermal shutdown.*

*The full CUBX (not -E or -L) and TUSL2 motherboards are so idiotic that they do not use the +3.3V bus from the ATX power supply at all. The 3.3V is generated locally on the board from +5V rail.
Simple boards with SDRAM, which have no voltage regulation on them, simply connect the memory power supply to the 3.3V bus of the PSU’s ATX power supply.

They seem to have a separate voltage regulator because they are pretty much the only boards with jumper settings for dimm and northbridge voltage (in case you need extra stability for high fsbs).

subhuman@xgtx wrote on Yesterday, 00:10:

shevalier wrote on 2026-04-23, 08:35:

Isn't this PSU actually an EOL? And then there’s the cooling. For some people, with three 120mm fans in the case, it’ll manage e […]
Show full quote

Mike_ wrote on 2026-04-23, 07:55:

It's not quite *that* bad. For aforementioned RM550x 3,3V/5V outputs have combined total power of 130W and from RM750x up it's 150W. It's not optimal, but should be usable for most old ATX setups.

https://assets.corsair.com/image/upload/corsa … 2018_Manual.pdf

Isn't this PSU actually an EOL?
And then there’s the cooling.
For some people, with three 120mm fans in the case, it’ll manage even with an ASUS TUSL2-C or CUBX.
But for others, even with a Duron, PSU go into thermal shutdown.*

*The full CUBX (not -E or -L) and TUSL2 motherboards are so idiotic that they do not use the +3.3V bus from the ATX power supply at all. The 3.3V is generated locally on the board from +5V rail.
Simple boards with SDRAM, which have no voltage regulation on them, simply connect the memory power supply to the 3.3V bus of the PSU’s ATX power supply.

They seem to have a separate voltage regulator because they are pretty much the only boards with jumper settings for dimm and northbridge voltage (in case you need extra stability for high fsbs).

Unfortunately, the voltage regulation is not separate.
In other words, on the same TUSL-2C, where the Intel I815 is powered by 1.8 volts, the AGP power supply could easily be connected to the PSU’s power output. Alas, the built-in converter affects the entire board at once.

tehsiggi wrote on 2026-05-27, 18:03:

for good measure.

The only problem is actually taking those "measurements".
For example, a 12V rail.
Ripple noise is less than 0.12V at rated current.
We apply a load using resistors and connect an oscilloscope. It’s a bit of a hassle, but feasible for around 100W. (But not for 500W+)
Load response/transient.
1.2V (1.5V for ATX 3.0 and newer) with a load change (from memory) of 20 to 80% of rated power.
Without a programmable electronic load – almost impossible.
If you measure it live on a PC, you’ll see a band of noise several volts wide. And try to figure out what it is. 🙁

shevalier wrote on Yesterday, 03:43:

Unfortunately, the voltage regulation is not separate. In other words, on the same TUSL-2C, where the Intel I815 is powered by 1 […]
Show full quote

subhuman@xgtx wrote on Yesterday, 00:10:

shevalier wrote on 2026-04-23, 08:35:

Isn't this PSU actually an EOL? And then there’s the cooling. For some people, with three 120mm fans in the case, it’ll manage e […]
Show full quote

Isn't this PSU actually an EOL?
And then there’s the cooling.
For some people, with three 120mm fans in the case, it’ll manage even with an ASUS TUSL2-C or CUBX.
But for others, even with a Duron, PSU go into thermal shutdown.*

*The full CUBX (not -E or -L) and TUSL2 motherboards are so idiotic that they do not use the +3.3V bus from the ATX power supply at all. The 3.3V is generated locally on the board from +5V rail.
Simple boards with SDRAM, which have no voltage regulation on them, simply connect the memory power supply to the 3.3V bus of the PSU’s ATX power supply.

They seem to have a separate voltage regulator because they are pretty much the only boards with jumper settings for dimm and northbridge voltage (in case you need extra stability for high fsbs).

Unfortunately, the voltage regulation is not separate.
In other words, on the same TUSL-2C, where the Intel I815 is powered by 1.8 volts, the AGP power supply could easily be connected to the PSU’s power output. Alas, the built-in converter affects the entire board at once.

tehsiggi wrote on 2026-05-27, 18:03:

for good measure.

The only problem is actually taking those "measurements".
For example, a 12V rail.
Ripple noise is less than 0.12V at rated current.
We apply a load using resistors and connect an oscilloscope. It’s a bit of a hassle, but feasible for around 100W. (But not for 500W+)
Load response/transient.
1.2V (1.5V for ATX 3.0 and newer) with a load change (from memory) of 20 to 80% of rated power.
Without a programmable electronic load – almost impossible.
If you measure it live on a PC, you’ll see a band of noise several volts wide. And try to figure out what it is. 🙁

I have a self-built controllable load that i use for such measurements. From my audio days.

The PSU is doing fine so far, but i don't want to leave the stinky HEC caps in if i don't have to.

tehsiggi wrote on Yesterday, 04:02:

shevalier wrote on Yesterday, 03:43:

Unfortunately, the voltage regulation is not separate. In other words, on the same TUSL-2C, where the Intel I815 is powered by 1 […]
Show full quote

subhuman@xgtx wrote on Yesterday, 00:10:

They seem to have a separate voltage regulator because they are pretty much the only boards with jumper settings for dimm and northbridge voltage (in case you need extra stability for high fsbs).

Unfortunately, the voltage regulation is not separate.
In other words, on the same TUSL-2C, where the Intel I815 is powered by 1.8 volts, the AGP power supply could easily be connected to the PSU’s power output. Alas, the built-in converter affects the entire board at once.

tehsiggi wrote on 2026-05-27, 18:03:

for good measure.

The only problem is actually taking those "measurements".
For example, a 12V rail.
Ripple noise is less than 0.12V at rated current.
We apply a load using resistors and connect an oscilloscope. It’s a bit of a hassle, but feasible for around 100W. (But not for 500W+)
Load response/transient.
1.2V (1.5V for ATX 3.0 and newer) with a load change (from memory) of 20 to 80% of rated power.
Without a programmable electronic load – almost impossible.
If you measure it live on a PC, you’ll see a band of noise several volts wide. And try to figure out what it is. 🙁

I have a self-built controllable load that i use for such measurements. From my audio days.

The PSU is doing fine so far, but i don't want to leave the stinky HEC caps in if i don't have to.

It would be interesting to see the results of this Codegen test under dynamic load

shevalier wrote on Yesterday, 04:43:

tehsiggi wrote on Yesterday, 04:02:

shevalier wrote on Yesterday, 03:43:

Unfortunately, the voltage regulation is not separate. In other words, on the same TUSL-2C, where the Intel I815 is powered by 1 […]
Show full quote

Unfortunately, the voltage regulation is not separate.
In other words, on the same TUSL-2C, where the Intel I815 is powered by 1.8 volts, the AGP power supply could easily be connected to the PSU’s power output. Alas, the built-in converter affects the entire board at once.

The only problem is actually taking those "measurements".
For example, a 12V rail.
Ripple noise is less than 0.12V at rated current.
We apply a load using resistors and connect an oscilloscope. It’s a bit of a hassle, but feasible for around 100W. (But not for 500W+)
Load response/transient.
1.2V (1.5V for ATX 3.0 and newer) with a load change (from memory) of 20 to 80% of rated power.
Without a programmable electronic load – almost impossible.
If you measure it live on a PC, you’ll see a band of noise several volts wide. And try to figure out what it is. 🙁

I have a self-built controllable load that i use for such measurements. From my audio days.

The PSU is doing fine so far, but i don't want to leave the stinky HEC caps in if i don't have to.

It would be interesting to see the results of this Codegen test under dynamic load

So much work stacks up.. If I get to it, I'll post it here 😀

tehsiggi wrote on Yesterday, 05:20:

So much work stacks up.. If I get to it, I'll post it here 😀

I have a fully functional IBM hard drive from the AVVER series (60GB).
During random access, the read/write heads move so erratically that they generate interference in the power supply of around 0.5V peak-to-peak.
It will be interesting to take a look at your Codegen. We look forward to it then.

shevalier wrote on 2026-05-24, 11:55:

From personal experience. […]
Show full quote

momaka wrote on 2026-05-19, 14:05:

Therefore, DO NOT use polymers.

From personal experience.

The FSP ATX-350PAF power supplies work perfectly with Capson LZ 1800 μF × 16 V × 13 mΩ capacitors (electrolytic capacitors with ultra-low ESR). These capacitors are not of particularly high quality in themselves, but their ESR is comparable to the figures in the datasheet, at least when it comes to new and genuine units.
There were no other 10mm diameter capacitors available. 🙁

This general-purpose polymer capacitor from Kemet has, in principle, very similar technical specifications. (10mOhm)
https://eu.mouser.com/datasheet/3/72/1/A750MW … 28M1CAAE010.pdf
I don't have capacitors of that value on hand, so I haven't specifically tested them.

The experiments were carried out on a Hipro HP-D5201AW PSU 520 W (APFC, 2 transistors forward , Champion CM6800TX PWM controller, group regulation).
I tried general use polymer caps 1500 µF 10V on the +5V rail and 2200 µF 6.3V caps on the +3.3V rail, and they work perfectly.
The 820 µF*16V capacitors work fine, but their capacitance is critically insufficient.
The power supply stability issues started precisely when I added five more capacitors to the two 820 µF capacitors to achieve the standard capacitance of 2200 µF*2.
Polymer capacitors have a very weak dependence of their ESR on capacitance. This means you can't replace 2200 µF (12 mOhm) with 3 x 680 µF, because 680 µF have an ESR of 13 or 14 mOhm, not 36 mOhm.

So, most likely, if you use the same number of polymer cap as electrolytic cap and the same capacity, there shouldn't be any problems.
But experimentation is necessary.

In parallel with the group stabilisation choke, a board containing 5 × 820 μF polymer capacitors has been added (which are covered by wires) to the 2 × 820 μF capacitors already installed in the power supply unit.

Let me tell you what. It seems that, with the help of Google’s Gemini AI, I’ve actually managed to set up the frequency compensation for the feedback loop.
I certainly wouldn’t have managed it on my own. 🙁
I can’t test it under a normal load, as thats only got 2 × 100 μF input capacitors left capacity(they used to be 120 μF back in the day, Hipro has achieved a significant reduction in costs).
I’d like to fit 170–220 μF * 400 V, but I can’t find any with an 18 mm diameter.

PS

I’d like to fit 170–220 μF * 400 V, but I can’t find any with an 18 mm diameter.

I found a Rubicon CXW 180*450μF for $2.
Man, why don't Chinese manufacturers just create their own brand of HCC - “honest Chinese capacitors”?
Something like, “Average quality, reasonable price—here's the datasheet.”
We’ve learnt how to make capacitors and promise they meet the datasheet specifications.

momaka wrote on 2026-04-27, 22:21:

Much better unit!
Only thing that worries me in that one is how long the "TL" primary cap will last. With APFC and it being rated only 400V *AND* being a rather low-quality brand... I just hope it goes out "gracefully", by loosing capacitance slow enough that the APFC circuit starts shutting down the PSU rather than MOSFETs going "kaboom".

momaka wrote on 2026-04-27, 22:21:

a) Any Japanese caps are left alone / *not* replaced. The only exception is if there are series in there that are well-known to be problematic (e.g. Nichicon PR). And for PSUs with APFC, the primary cap gets pulled out and checked for capacitance, ESR, and leakage current under full voltage (despite being a Japanese brand), because as I mentioned, the APFC circuit in some PSUs really runs them too hard.

Btw, I recapped my Enermax, which apparently is active PFC so would it be a good idea to replace the primary cap? I don't recognize its brand, either. I happen to have a couple of these chemi-con SMQ caps available if they would be suitable.

Mike_ wrote on Today, 11:54:

Btw, I recapped my Enermax, which apparently is active PFC so would it be a good idea to replace the primary cap? I don't recogn […]
Show full quote

momaka wrote on 2026-04-27, 22:21:

Much better unit!
Only thing that worries me in that one is how long the "TL" primary cap will last. With APFC and it being rated only 400V *AND* being a rather low-quality brand... I just hope it goes out "gracefully", by loosing capacitance slow enough that the APFC circuit starts shutting down the PSU rather than MOSFETs going "kaboom".

momaka wrote on 2026-04-27, 22:21:

a) Any Japanese caps are left alone / *not* replaced. The only exception is if there are series in there that are well-known to be problematic (e.g. Nichicon PR). And for PSUs with APFC, the primary cap gets pulled out and checked for capacitance, ESR, and leakage current under full voltage (despite being a Japanese brand), because as I mentioned, the APFC circuit in some PSUs really runs them too hard.

Btw, I recapped my Enermax, which apparently is active PFC so would it be a good idea to replace the primary cap? I don't recognize its brand, either. I happen to have a couple of these chemi-con SMQ caps available if they would be suitable.

The attachment primary.jpg is no longer available

It's an Hitachi AIC capacitor, imho not worth changing.

Mike_ wrote on Today, 11:54:

Btw, I recapped my Enermax, which apparently is active PFC so would it be a good idea to replace the primary cap? I don't recognize its brand, either. I happen to have a couple of these chemi-con SMQ caps available if they would be suitable.

The attachment primary.jpg is no longer available

In PSU without an APFC, the main converter itself is the only high-frequency load.
The APFC is essentially a step-up converter.
Thus, this capacitor is sandwiched between two high-current sources of high-frequency interference, and it often ends up in a bad state, unlike in older topologies.
When it fails (due to capacitance and ESR), the APFC (MOSFETs, diode, and often the controller) goes burn, making the repair of such a PSU cost half as much as buying a new one.
The primary capacitors in this topology fail completely without any outward signs.

If you’ve taken apart a PSU with an APFC, you always need to carry out 2½ steps:
- Measure the voltage across the large capacitor. For a 400V capacitor, the reading should be 380V ± 2–3V (for other ratings, check the voltage dividers connected to the controller. However, it definitely cannot be less than 375 volts)
- Measurement of ESR
- Measurement of capacitance, but this will be calculated automatically from the previous step.
According to current trends, 1 watt ~ 1 μF.

P.S. Alas, this is now one of the most likely points of failure. 🙁
We need to check them and wait for the price of these components to drop tenfold.
https://eu.mouser.com/ProductDetail/Vishay-Ro … kERvpWojg%3D%3D

shevalier wrote on Today, 13:28:

In PSU without an APFC, the main converter itself is the only high-frequency load. The APFC is essentially a step-up converter. […]
Show full quote

In PSU without an APFC, the main converter itself is the only high-frequency load.
The APFC is essentially a step-up converter.
Thus, this capacitor is sandwiched between two high-current sources of high-frequency interference, and it often ends up in a bad state, unlike in older topologies.
When it fails (due to capacitance and ESR), the APFC (MOSFETs, diode, and often the controller) goes burn, making the repair of such a PSU cost half as much as buying a new one.
The primary capacitors in this topology fail completely without any outward signs.

If you’ve taken apart a PSU with an APFC, you always need to carry out 2½ steps:
- Measure the voltage across the large capacitor. For a 400V capacitor, the reading should be 380V ± 2–3V (for other ratings, check the voltage dividers connected to the controller. However, it definitely cannot be less than 375 volts)
- Measurement of ESR
- Measurement of capacitance, but this will be calculated automatically from the previous step.
According to current trends, 1 watt ~ 1 μF.

P.S. Alas, this is now one of the most likely points of failure. 🙁
We need to check them and wait for the price of these components to drop tenfold.
https://eu.mouser.com/ProductDetail/Vishay-Ro … kERvpWojg%3D%3D

I don't have an ESR meter, so that's out of the question...