AeroCool VP-550 : IC identification issue

[bendeg]

Re: AeroCool VP-550 : IC identification issue

Final test to be certain :

A PC powered with ATX and CPU connectors only (I didn't remove the PC power supply to put the repaired one) => ok, no magic smoke or magic smell, fan is spinning gently.

It is working fine under normal load (CPU, system fan, graphic card and its fan, etc...)

[CapLeaker]

Re: AeroCool VP-550 : IC identification issue

Congrats! Well done on your first serious electronics repair and you did good!

[bendeg]

Re: AeroCool VP-550 : IC identification issue

2nd test, fan connected, with light bulb + power supply tester :
power on (all voltages seemed to be good on power supply tester), fan on, then all went off...

3rd test, no light bulb, power supply tester : all voltages ok (see attachment)

4th test , no light bulb, with 1 HDD : ok, all seems working so far !

As this is my very first "serious" electronic repair, I'm very happy !

Thanks so much guys for all your help !

[bendeg]

Re: AeroCool VP-550 : IC identification issue

So far, so good :

New components soldered :

• fuse (6.3A)
• thermistor (SCK-2R56, 2.5 ohms, 6A)
• bridge rectifier (GBU810 instead of GBU606)
• mosfet (MDP13N50)
• diode (original DP8S60P was ok but replaced with a MUR1560)
• capacitor (180uF, 450V, Nichicon). No additionary capacitor as advised by momaka...
• PWM STR6069H

1st test with 80W light bulb in series looks ok, fuse doesn't blow ! (fan not connected yet)

• Small flash then off
• Stand by voltage (purple wire) : 5.22V
• Green wire : 2.65V

Next step : light bulb + power on (green on black) ?

[bendeg]

Re: AeroCool VP-550 : IC identification issue

Ok, you confirm, thanks !

Yup, I will confirm what you have is a 4-band resistor with the 5th band being used for temp coefficient.
So looks like you found the right info.

In terms of the power rating, though, that looks like either a 2-Watt or 3-Watt resistor to me. 5 Watt resistors tend to be HUGE, so I don't think this one is it.

Probably safe to go with a 3-Watt... or better yet, don't replace it if it reads OK / low-resistance.

Also note, these are flame-proof metal film resistors. Don't use carbon composite unless you want to see a small inferno in there (if/when something fails anyways.)


Probably.
Quite possible the info there hasn't been update in years.

Thank you for that information !

I've sent a message to the guy who created the "The Resistor FAQ" thread. I can't do better.

[momaka]

Re: AeroCool VP-550 : IC identification issue

I think I finally found a good info here, at the end of the page, it tells about some exceptions to the rule :

https://eepower.com/resistor-guide/r...or-color-code/



So, in my case :

Bands 1 and 2 : BROWN - BLACK => 10
Band 3 : SILVER (multiplication factor) => 10 * 0.01 = 0.1 Ohms
Band 4 : GOLD (tolerance) => 5%
Band 5 : GREEN (temperature coefficient) => 20 ppm/°C

Now, that said, it doesn't tell about resistor wattage...

There's another 5 bands resistor in the vicinity of the first one, same look but a bit smaller : RED - RED - SILVER - GOLD - GREEN
=> 0.22 ohms and that's indeed what my DMM gives me.

Yup, I will confirm what you have is a 4-band resistor with the 5th band being used for temp coefficient.
So looks like you found the right info.

In terms of the power rating, though, that looks like either a 2-Watt or 3-Watt resistor to me. 5 Watt resistors tend to be HUGE, so I don't think this one is it.

Probably safe to go with a 3-Watt... or better yet, don't replace it if it reads OK / low-resistance.

Also note, these are flame-proof metal film resistors. Don't use carbon composite unless you want to see a small inferno in there (if/when something fails anyways.)

Maybe it would be a good idea to tell about that exception in the FAQ about resistors.

Probably.
Quite possible the info there hasn't been update in years.

[bendeg]

Re: AeroCool VP-550 : IC identification issue

I think I finally found a good info here, at the end of the page, it tells about some exceptions to the rule :

https://eepower.com/resistor-guide/r...or-color-code/

5 band resistor with a 4th band of gold or silver

Five band resistors with a fourth band of gold or silver form an ]exception and are used on specialized and older resistors. The first two bands represent the significant digits, the 3rd is the multiplication factor, the 4th is the tolerance, and the 5th is the temperature coefficient (ppm/˚C).

So, in my case :

Bands 1 and 2 : BROWN - BLACK => 10
Band 3 : SILVER (multiplication factor) => 10 * 0.01 = 0.1 Ohms
Band 4 : GOLD (tolerance) => 5%
Band 5 : GREEN (temperature coefficient) => 20 ppm/°C

Now, that said, it doesn't tell about resistor wattage...

There's another 5 bands resistor in the vicinity of the first one, same look but a bit smaller : RED - RED - SILVER - GOLD - GREEN
=> 0.22 ohms and that's indeed what my DMM gives me.

Maybe it would be a good idea to tell about that exception in the FAQ about resistors.

[CapLeaker]

Re: AeroCool VP-550 : IC identification issue

What is the power rate of that 5 bands resistor (1 stage, near the big capacitor) ?

I couldn't find a color code calculator that found the answer using 5 bands.
Looks like it's a 4 bands resistor with a fifth one...

The power rate in Watts of a resistor is based on the size (length and diameter).
And yes there are 5 band colour code resistor calculators. Usually large Chinese resistors have weird values and are harder to get. In a nut shell the 3 and 4 colour band resistors are not telling the power rating in watts. It just tells you the resistance, multiplier and tolerance in that order. The 5th band is temperature coefficient. It’s still morning… hope it makes sense. Lol

[bendeg]

Re: AeroCool VP-550 : IC identification issue

What is the power rate of that 5 bands resistor (1 stage, near the big capacitor) ?

I couldn't find a color code calculator that found the answer using 5 bands.
Looks like it's a 4 bands resistor with a fifth one...

According to its color, BROWN - BLACK - SILVER - GOLD - GREEN...
(correct me if I'm wrong, I'm color blind, I've got problems especially with the greens, I'm a "deuteranope" guy)

...it's a 4 bands resistor, and it should be a 0.1 Ohm 5% resistor, right ? (and indeed, I can measure it with my DMM).

Then, if it's a 4 bands resistor, what is the meaning of the last green band ?

Also, how can I find the power rating so I could order one if needed (even if it looks in good shape to my eyes) ?
Dimensions : width = ~17mm, diameter = ~6mm
Is it a 5W resistor ? More than 5W ?

[bendeg]

Re: AeroCool VP-550 : IC identification issue

...

Oh, that's a lot of information.

There's more to do now...I'll update this thread when I've got all the components.

Thank you very much !!!

[momaka]

Re: AeroCool VP-550 : IC identification issue

1)

Fortunately, STR-6069H hasn't been delivered yet. Thank you for the advice !

And indeed, as I can read in the datasheet, page 23, for pin 7 (I believe this is what you want to tell me...) :
"
(7) Thermal Considerations
Because the power MOSFET has a positive thermal
coefficient of R DS(ON) , consider it in thermal design.
Since the copper area under the IC and the D/ST pin
trace act as a heatsink, its traces should be as wide as
possible.
"

Yes, this is exactly the reason.

2)
I already planned to change the main cap (JunFu, 180uF, 400V, 85°, LG, 25mm x 40mm snap-in) : my DMM shows only 5.5uF instead of 180uF...

I have 2 different new caps replacement:

• Nichicon, 180uF, 400V, 105*, PT(M), 18mm x 40mm
• Nichicon, 180uF, 450V, 105°, GG(M), 25mm x 40mm snap-in and fits perfectly (same size as original)

Is this a problem to put the first one (18mm x 40mm) ? Should the size match the original perfectly ?

Use the 450V cap.

For the primary cap, bigger is always better - both in terms of size and capacity. This is one part where manufacturers save the most money by using parts that are just adequate enough for the power rating of the PSU at the expense of the part not lasting as long.

Since the 450V cap fits perfectly in terms of size, use that. It should last a lot longer than the original.
Better yet, if you can get an bigger cap (say, 220 uF) to fit and if it doesn't cost significantly more, that would be nice too. But don't sweat it too much. Looks like you already found a cap that fits and rated for a higher voltage, which is even better.

I try not to use anything lower than 420V -rated caps for PSUs with APFC circuits. 400V caps just seem to not last in certain APFC circuits.

3)
polypropylene cap : you mean a capacitor that looks like a yellow X2 capacitor or that looks more like a woman handbag, a brown purse like this one :

Use the "woman handbag" one... and LOL! That's actually a pretty accurate visual description, I like it.
Just make sure it is Polypropylene (PP) type. Polyester (PE) can look the same... and for the most part, it's still a film capacitor and will work mostly the same. However, PP has better frequency response and is the one recommended for high-frequency circuits.

I don't know if I can recommend X2, since it's mostly for AC input filter applications, and here on the APFC we are dealing with high-voltage DC (with small HF AC ripple.)

4)
You mean that if a test with a 60-100W bulb in series pass with the new IC, I should try with 200-300W bulb afterwards or an hair dryer ??? Not sure I understand that part, do you mean connecting a hair dryer on the "low" setting in series with the PSU ?

Install new IC and test with 60-100W bulb in series with the PSU, like this:
https://www.badcaps.net/forum/showpo...51&postcount=6
https://www.badcaps.net/forum/showpo...4&postcount=70

The 60-100W bulb is only when you want to test if the PSU outputs 5VSB. Once you want to test the entire PSU (by connecting PS_ON to ground), you will need to replace the 60-100W bulb with something more powerful, because of the APFC circuit. APFC does not like high-impedance paths on it's supply input, which is what the 60-100W will be at the high frequency that the APFC circuit pulses its MOSFETs to primary ground. What will likely happen with the 60-100W bulb if you try to power-on the PSU (PS_ON to ground) is the bulb will likely flash on-and-off very quickly or just stay lit continuously and not let the PSU start.

For this reason, you connect a higher-power linear device, such as hair dryer, toaster, heat gun (non-digital one)... or just 4+ 60-100 Watt incandescent or halogen bulbs in parallel (4x 60W bulbs in parallel = 240 Watts equivalent.)

I don't have a hair dryer but I have a small 500-600W hand mixer (in the kitchen) with a turbo button (a "normal" button and a "turbo" button), that might work as well :-D

Mixer probably won't work, since it has a motor inside. You want a device / appliance with some sort of heating element and simple On-Off switch control (i.e. no digital circuits or "soft" push-buttons.)

If you don't have such appliance, just use incandescent or halogen bulbs. I know that in EU it's harder to get these now. If that's the case and you can't find such higher-power bulbs, use 40W incandescent "appliance" type bulbs - these are regular incandescent bulbs, but they work the same. EU can't ban these, because appliances such as your kitchen oven can't use LED or CFL bulbs inside it - they simply won't last. So "appliance" incandescent bulbs are exempt from the incandescent bulb ban.

I knew for the test with a bulb of about 60-100W, but not more, why is that needed ? To test little by little to make sure nothing wrong happens ?

Pretty much, yes.
60-100W for testing the 5VSB.
200-300W or more for testing whole PSU (PS_ON connected to ground.)

That being said, beware that some PSUs won't work without a small load on the output. You may have to connect an HDD or two or three... or a 12V halogen bulb on the 12V/5V rails. In such case, however, just know that the higher the load you put on the output of the PSU, the bigger the power rating of the series incandescent/halogen bulb will need to be on the input.

IME, the limit is usually around 20-25% of the power of the series bulbs - i.e., if you use a 300W halogen bulb on the input of the PSU, you may be able to pull only about 60-75W of power on the output. This is for PSUs without APFC. With APFC, the dynamic can change a little... but assuming 20-25% again is a pretty safe bet.

[bendeg]

Re: AeroCool VP-550 : IC identification issue

Don't!

The Source and Drain pins on the IC (mostly the Drain pins) are used as a heatsink. If you install on a socket, the chip will likely overheat and die faster.

Solder it in, instead.

Make sure to also replace that failed primary cap. This is likely the reason the IC blew in the first place. With APFC circuits, if the main cap fails, a nasty high-voltage spike will appear downstream on the primary side. Sometimes it kills nothing and PSU just blows the fuse on the input. Other times, it can kill the APFC MOSFETs, main PS MOSFETs, or even the 5VSB / standby offline switch IC (STR-A6069H in your case.)

My personal suggestion is to also put a polypropylene cap in parallel with the mains cap, so in case of a failure of the big cap again, the PP cap can take some of the energy / high-voltage spike and be less likely to end up with other dead components. Use a PP cap between 0.1 to a few uF, rated for 450V minimum (630V better.)

Lastly, use a dim bulb tester when testing the 5VSB with the replaced IC. 60-100 Watts incandescent or halogen will do. Once 5VSB function is verified, you can test the PSU with a slightly bigger halogen bulb in series - 200-300 Watts... or hair dryer on the "low" setting.

1)

Fortunately, STR-6069H hasn't been delivered yet. Thank you for the advice !

And indeed, as I can read in the datasheet, page 23, for pin 7 (I believe this is what you want to tell me...) :
"
(7) Thermal Considerations
Because the power MOSFET has a positive thermal
coefficient of R DS(ON) , consider it in thermal design.
Since the copper area under the IC and the D/ST pin
trace act as a heatsink, its traces should be as wide as
possible.
"

2)
I already planned to change the main cap (JunFu, 180uF, 400V, 85°, LG, 25mm x 40mm snap-in) : my DMM shows only 5.5uF instead of 180uF...

I have 2 different new caps replacement:

• Nichicon, 180uF, 400V, 105*, PT(M), 18mm x 40mm
• Nichicon, 180uF, 450V, 105°, GG(M), 25mm x 40mm snap-in and fits perfectly (same size as original)

Is this a problem to put the first one (18mm x 40mm) ? Should the size match the original perfectly ?

3)
polypropylene cap : you mean a capacitor that looks like a yellow X2 capacitor or that looks more like a woman handbag, a brown purse like this one :

4)
You mean that if a test with a 60-100W bulb in series pass with the new IC, I should try with 200-300W bulb afterwards or an hair dryer ??? Not sure I understand that part, do you mean connecting a hair dryer on the "low" setting in series with the PSU ?
I don't have a hair dryer but I have a small 500-600W hand mixer (in the kitchen) with a turbo button (a "normal" button and a "turbo" button), that might work as well :-D

I knew for the test with a bulb of about 60-100W, but not more, why is that needed ? To test little by little to make sure nothing wrong happens ?

[momaka]

Re: AeroCool VP-550 : IC identification issue

I also plan to install a 8 pins socket so IC can be replaced easily if needed.

Don't!

The Source and Drain pins on the IC (mostly the Drain pins) are used as a heatsink. If you install on a socket, the chip will likely overheat and die faster.

Solder it in, instead.

Make sure to also replace that failed primary cap. This is likely the reason the IC blew in the first place. With APFC circuits, if the main cap fails, a nasty high-voltage spike will appear downstream on the primary side. Sometimes it kills nothing and PSU just blows the fuse on the input. Other times, it can kill the APFC MOSFETs, main PS MOSFETs, or even the 5VSB / standby offline switch IC (STR-A6069H in your case.)

My personal suggestion is to also put a polypropylene cap in parallel with the mains cap, so in case of a failure of the big cap again, the PP cap can take some of the energy / high-voltage spike and be less likely to end up with other dead components. Use a PP cap between 0.1 to a few uF, rated for 450V minimum (630V better.)

Lastly, use a dim bulb tester when testing the 5VSB with the replaced IC. 60-100 Watts incandescent or halogen will do. Once 5VSB function is verified, you can test the PSU with a slightly bigger halogen bulb in series - 200-300 Watts... or hair dryer on the "low" setting.

[bendeg]

Re: AeroCool VP-550 : IC identification issue

By not using Google as a search engine. Yw.!
Using a 6069H or a 6059H doesn't matter. The Difference is the max input voltage handling of the IC.

Indeed, I only tried google search...

Oh, ok, it'll be the A6069H or any A606x then (700V max), not the A6059H (650V max).

Thank you very much for that info !

[CapLeaker]

Re: AeroCool VP-550 : IC identification issue

As I see in the datasheet A6059H has the same specs (it would probably be a good replacement IC)

But how the hell did you find that !?!?

Thank you so much !!!

By not using Google as a search engine. Yw.!
Using a 6069H or a 6059H doesn’t matter. The Difference is the max input voltage handling of the IC.

[bendeg]

Re: AeroCool VP-550 : IC identification issue

Yes, in these series is STR-A6069H.

It's a slightly different configuration (there's a second capacitor near the heatsink and there's no what looks like a yellow capacitor close to pin 3 and 4)

But anyway, thank you so much !!!

I also plan to install a 8 pins socket so IC can be replaced easily if needed.

Keep you all updated once I've got all parts soldered.
Fingers crossed...

[bendeg]

Re: AeroCool VP-550 : IC identification issue

I did some digging (being bored this morning ) and I found one single reference on the AeroCool VP-550 that chip in question being an STR-A6069H.

Good luck!

As I see in the datasheet A6059H has the same specs (it would probably be a good replacement IC)

But how the hell did you find that !?!?

Thank you so much !!!

[lotas]

Re: AeroCool VP-550 : IC identification issue

Yes, in these series is STR-A6069H.

[CapLeaker]

Re: AeroCool VP-550 : IC identification issue

I did some digging (being bored this morning ) and I found one single reference on the AeroCool VP-550 that chip in question being an STR-A6069H.

Good luck!

[bendeg]

Re: AeroCool VP-550 : IC identification issue

Hopefully you were careful opening the case. The rest of the lid of the IC should be floating around the inside of the power supply. I think you are right on it being a Sanken STR-A60xx series. Luckily there aren't many to choose of. However these very few to choose of have different switching frequency. 67 vs 100khz.

I've received the STR-A6059H (working at 100kHz), and ready to be soldered, but I think it may not be a good idea (I've ordered 6 others of the series, mainly those that works at 67kHz, but 2 others that works at 100kHz as well, a bit crazy but I must try...) :

Let's suppose the proper IC should work at 67kHz, it such a case it means it also expects modulation feedback that would not match if I put one that works at 100kHz, right ?

In a nutshell, in such a case, the "modulation time" of the feedback would be too long with the STR-A6059H and the IC would work almost twice as much time, maybe at 100%, overwhelming the circuit. See what I mean ?

It would be the other way round if I put a 67kHz one intead of a chip that should work at 100kHz ?

Am I wrong ? Or is the feedback automatically "sensed" ?