Allied AL-8460BTX (450W) worth recapping?

[goodpsusearch]

This power supply uses the same layout as those Solytech psus, reviewed by c_hegge:
http://hardwareinsights.com/wp/xigma...rp-vc503-500w/
http://hardwareinsights.com/wp/the-2...ply-roundup/5/
http://hardwareinsights.com/wp/the-2...ply-roundup/5/

They all made it to 350-400W with very good voltage regulation and low ripple output except the Xigmatec's 3.3V.

The Allied has full input filter, including MOV protection, 2x 470uF 200V primary caps, 2xTT2194 NPN transistors and FTA02N65 mosfet for 5vsb.

There is passive pfc coil, not fake.

The main transformer is true ERL-35 and the 5vsb transformer is EEL-19.

Secondary:

PWM controller is chip of the year 2005Z.

As you can see, this psu has a fan controller board attached to the secondary heatsink.

5vsb: 1xSR560 2x1000uF 10V YC TK

3.3V: 1xSBL2045CT 1x2200uF 10V Koshin KLH, 1x1000uF 10V YC TK

5V: 2xSBL2045CT (total 40A) 1x2200uF 10V Koshin KLH, 1x1000uF 10V YC TK

12V: 1xSB3060PT 30A, 2x1000uF 16V YC TK

-12V: 1x470uF 16V YC TK

There are pi coils installed in every single output including 5vsb and negative voltage rail -12V.

Do you think it is worth recapping?


Btw the bleeder resistors are not of the same value:
R3: 462kohm
R4: 330kohm
so, the rectified voltage is not divided equally. I have no idea why they do this.

[stj]

Re: Allied AL-8460BTX (450W) worth recapping?

it looks pretty good.
only thing i'd do is ditch that small pcb with the fan-crippler on it - connect the fan to 12v.

btw, what type of fan vent is it?
cutouts, or a proper wire guard?
if it's just cutouts then jigsaw or dremel it out & fit a real guard.
you will move more air & make less noise.

[goodpsusearch]

Re: Allied AL-8460BTX (450W) worth recapping?

Cutous, but not like this:
http://hardwareinsights.com/wp/wp-co...013/02/PSI.jpg

It's like this:
http://www.atxpowersupplies.com/imag...00-60SPV-D.gif

[c_hegge]

Re: Allied AL-8460BTX (450W) worth recapping?

only thing i'd do is ditch that small pcb with the fan-crippler on it - connect the fan to 12v.

Nah. You don't want it wearing the fan and sucking up more dust than it really needs to. I'd just give it a full re-cap and treat it as a 300W unit, and it should do OK.

EDIT: You could also bend the bottom row of fins downwards somewhat. That might give the coils and caps a bit more air.

[stj]

Re: Allied AL-8460BTX (450W) worth recapping?

cut that metal out for the fan.
and remember, your not just cooling the psu, your pulling the hot air out of the case usually.

[goodpsusearch]

Re: Allied AL-8460BTX (450W) worth recapping?

Some news..

The voltage drop on the primary cap in parallel with 330kohm is 140V.
The other capacitor is in parallel with 462kohm bleeding resistor and the voltage measured is 196V!

This is too close to the capacitor limit.

Either I have to use 220V/250V capacitors, or change the bleeding resistors. Those resistors are there for a reason. Their asymmetry is part of the circuit design. Would you take the risk and put 2 resistors of the same value instead?

[mur]

Re: Allied AL-8460BTX (450W) worth recapping?

Is this psu made for 220v only? Because if you would use it on 110v power, both caps would get the same voltage anyway because of doubler circuit. If the voltage is that far asymmetrical I think you have bad primary caps.

[kc8adu]

Re: Allied AL-8460BTX (450W) worth recapping?

i suspect the bleeder resistor is an error.no reason to do that and is suboptimal unless you intend for the unit to die young.
when i build cap banks for amplifiers i use 1% tolerance if i can.lots of cap failures in these amplifiers were caused by drifting bleeders.if one opens that cap goes bang!

[goodpsusearch]

Re: Allied AL-8460BTX (450W) worth recapping?

Some news..

The voltage drop on the primary cap in parallel with 330kohm is 140V.
The other capacitor is in parallel with 462kohm bleeding resistor and the voltage measured is 196V!

This is too close to the capacitor limit.

Either I have to use 220V/250V capacitors, or change the bleeding resistors. Those resistors are there for a reason. Their asymmetry is part of the circuit design. Would you take the risk and put 2 resistors of the same value instead?

Please ignore that post. It turns out that the pcb trace connected to the + and the - leads of the 2 primary capacitors is part of the 5vsb circuit or something like that. That's the reason the 2 bleeding resistors have different values. One of the 2 primary caps stored DC voltage, part of it, is used by the circuit. Sorry for the bad English.


I replaced the TT2194 NPN transistors with FJA13009. I used mica pads and Arctic Silver 5.

I fully recapped the power supply with 100% Javanese capacitors.

I replaced the 470uF Koshin primaries with 800uF 200V Panasonic EE.

I used Chemicon LXZ and Nichicon PW 2200uF 16V caps for every main voltage output.

Finally, I soldered a 14.8Kohm resistor in parallel with thermal sensor to tweak the fan controller circuit to move more air.

It's a shame I don't have a psu load tester. How many watts do you think it can do now?

I also got to add an extra cable providing 1 SATA connector, cause the psu had none.

[everell]

Re: Allied AL-8460BTX (450W) worth recapping?

Looking at the model number, AL-8460BTX, The BTX at the end makes me wonder. I know that a BTX computer case has a different layout from the ATX computer case. Does anyone know what the differences are for the power supplies? Perhaps how it is mounted or which side the AC cord plugs into? Or perhaps the length of the cables connecting to the mother board, hard drive, and CD ROMs?

[dmill89]

Re: Allied AL-8460BTX (450W) worth recapping?

Looking at the model number, AL-8460BTX, The BTX at the end makes me wonder. I know that a BTX computer case has a different layout from the ATX computer case. Does anyone know what the differences are for the power supplies? Perhaps how it is mounted or which side the AC cord plugs into? Or perhaps the length of the cables connecting to the mother board, hard drive, and CD ROMs?

I once had a BTX computer, there are no electrical or physical-size diferences with the psu. However the cable lengths would likely be different due to the different layout of the case.

One thing worth noting with BTX the case was "reversed" with the motherboard on the left side of the case and tops of the expansion cards facing up. The primary advantage with BTX was cooling/airflow with the CPU and North-Bridge mounted near the center of the motherboard and inline with each other with front and rear fans on either side of these components (usually 90mm or 120 mm). Then the drive-cage was split with the optical drives on top and the HDDs below with the front fan between the two and right infront of the CPU. This helped tremendiously with cooling the very hot running "Netburst" chips, however once more efficent CPUs came out the extra cooling was no longer necessary and BTX quickly died off.

Example of BTX below:

[Pentium4]

Re: Allied AL-8460BTX (450W) worth recapping?

Cool PSU. Back before they exclusively used YC. Koshin caps are pretty terrible. Excellent job fixing it up, it looks good! I dare say it could do 375W with ripple well in spec, these do surprisingly well with ripple suppression, and i'm sure even better with the caps you used. Considering that smallish primary heatsink, those upgraded switchers would help quite a bit, that's usually what fails first on these old Deer PSU's.

[goodpsusearch]

Re: Allied AL-8460BTX (450W) worth recapping?

Thank you!

I forgot to mention that the 3 main voltages are on the high side, for example a reading gave 5.25V 12.3V and 3.4V.

[Th3_uN1Qu3]

Re: Allied AL-8460BTX (450W) worth recapping?

That is okay. Cheap PSUs are tuned to have their voltage output on the high side, because they KNOW it will drop under load...