Figured I’d separate the recapping details of this PSU in a different thread so that the information can be easier to find. To be more specific, this thread is about recapping of the Delta DPS-300AB-24b, shown here. All details about the build quality of that PSU are posted there.
Here is the PSU, just to see what we are dealing with:
And here is a cap diagram:
Let’s break it down what goes where…
* 5VSB
*** cap spot C952: Ltec LZG, 10V, 2200 uF, 10x30 mm before PI coil
*** cap spot C953: Taicon PW, 10V, 680 uF, 8x15 mm after PI coil (note: spot for this cap has 5 mm lead spacing and can accommodate a 10 mm diameter cap, if needed)
* 3.3V Rail
*** cap spot C302: Ltec LZG, 6.3V, 2200 uF, 10x25 mm before PI coil
*** cap spot C303: Ltec LZG, 6.3V, 2200 uF, 10x25 mm after PI coil
* 5V Rail
*** cap spot C103: Ltec LZG, 10V, 2200 uF, 10x30 mm before PI coil
*** cap spot C104: Ltec LZG, 10V, 2200 uF, 10x30 mm after PI coil
* 12V Rail
*** cap spot C153: United Chemicon KZH, 16V, 1000 uF, 8x20 mm (note: spot for this cap has 5 mm lead spacing and can accommodate a 10 mm diameter cap, if needed)
*** cap spot C152: Ltec LTG, 16V, 1500 uF, 10x20 mm
* -12V Rail
*** cap spot C252: Taicon PW, 25V, 220 uF, 8x13 mm before 7912 linear regulator
*** cap spot C253: Taicon PW, 25V, 100 uF, 6.3x11 mm after 7912 linear regulator
* “startup” capacitor (primary side)
*** Ltec LZG, 35V, 100 uF, 8x13 mm
Caps removed…
And here is the PSU in its recapped form:
For the 5VSB rail, I only changed the bad Ltec LZG, 10V, 2200 uF cap with a Panasonic FM, 6.3V, 2200 uF. The Panny FM is significantly better in terms of ESR and ripple current compared to the Ltec LZG (2470 mA and 0.018 Ohms impedance vs. 1920 mA and 0.030 Ohms impedance.) Deviating from the original PSU cap specs is usually not recommended. But in this case, it’s the 5VSB rail, which often does well even with very low ESR caps like Panasonic FM and FR, along with Rubycon ZLH.
Speaking of Rubycon ZLH, I used two 6.3V, 2200 uF of these to replace the 2x Ltec LZG, 6.3V, 2200 uF on the 3.3V rail. A bit overkill, given that the above ZLH are rated for 1960 mA and 0.020 Ohms impedance (vs. 1525 mA and 0.044 Ohms for the Ltec LZG.) But for a 3.3V rail with a mag-amp design, again very low ESR caps are usually not a problem.
The 5V rail received a different treatment: 2x Rubycon YXJ, 10V, 2200 uF. These are actually rated worse than the 2x Ltec LZG, 10V, 2200 uF caps they replaced in terms of RC and ESR (only 1400 mA and 0.046 Ohms.) Seems that I should have had these and the ZLH caps on the 3.3V rail swapped. But I knew this PSU would go into a modern PC with very little power draw on the 5V rail, so this slight “degradation” in cap specs will not matter. That said, for anyone doing a more than one of these Delta PSUs at the same time, you can just order 5x Rubycon ZLH 6.3V, 2200 uF caps to replace the ones on the 5V, 3.3V, and 5VSB rail with that. This can help with saving on caps when buying in bulk quantities of 10 or more online.
For the 12V rail, I swapped both caps with 2x Panasonic FM, 16V, 1200 uF, even though the United Chemicon KZH, 16V, 1000 uF cap is fine and of good quality. The only reason for that was because I didn’t have any 16V, 1500 uF caps for the Ltec LTG that aren’t motherboard grade ultra-low ESR. So going with the two Panny FM 1200 uF caps gave roughly the same capacitance (2400 uF total compared to 2500 uF before) and similar ESR specs. What I’d like to note here is that the Chemicon KZH 1000 uF cap is 8 mm diameter, but the holes on the PCB are for a cap with 5 mm lead spacing. Thus, using a 10 mm cap is possible – and that’s what I did with the Panny FM caps above. Also, both caps on the 12V rail are connected together without any PI coils or shunt resistors between them, so changing the capacitance of the two caps individually will not cause any imbalance (provided the total capacitance is about the same.)
And finally, for the -12V rail, I only swapped the Taicon PW, 25V, 220 uF cap with a Rubycon YXJ, 25V, 220 uF cap. The other cap filters the output of the 7912 linear regulator for the -12V rail, so it doesn’t really see any ripple or stress. Thus, I left it alone. Also, I tend to trust Taicon a little more compared to the other Taiwanese caps.
Oh, and for the “startup” cap, I don’t remember what I used, but it was likely in the low-ESR category of Nichicon PW and PM or similar. Same voltage and capacitance here, as always.
There are also a few other small caps in this PSU: If I remember correctly, these were 1x 50V, 1uF; 1x 50V, 4.7 uF; and 2x 25-50V (?), 10 uF (I can’t remember exactly, though.) Only mentioning this in case anyone wants to do a full PSU recap. I regularly keep a small stock of 1 uF, 2.2 uF, 4.7 uF, 10 uF, 22 uF, and 47 uF (typically 50V –rated) caps on hand for when I’m recapping PSUs, as these values tend to show up often.
And that completes this DPS-300AB-24b recap.
The PSU is already installed into a PC. I haven’t had to use that PC much, but every time I did, it’s booted reliably, like all of my other recapped PSUs.
Here is the PSU, just to see what we are dealing with:
And here is a cap diagram:
Let’s break it down what goes where…
* 5VSB
*** cap spot C952: Ltec LZG, 10V, 2200 uF, 10x30 mm before PI coil
*** cap spot C953: Taicon PW, 10V, 680 uF, 8x15 mm after PI coil (note: spot for this cap has 5 mm lead spacing and can accommodate a 10 mm diameter cap, if needed)
* 3.3V Rail
*** cap spot C302: Ltec LZG, 6.3V, 2200 uF, 10x25 mm before PI coil
*** cap spot C303: Ltec LZG, 6.3V, 2200 uF, 10x25 mm after PI coil
* 5V Rail
*** cap spot C103: Ltec LZG, 10V, 2200 uF, 10x30 mm before PI coil
*** cap spot C104: Ltec LZG, 10V, 2200 uF, 10x30 mm after PI coil
* 12V Rail
*** cap spot C153: United Chemicon KZH, 16V, 1000 uF, 8x20 mm (note: spot for this cap has 5 mm lead spacing and can accommodate a 10 mm diameter cap, if needed)
*** cap spot C152: Ltec LTG, 16V, 1500 uF, 10x20 mm
* -12V Rail
*** cap spot C252: Taicon PW, 25V, 220 uF, 8x13 mm before 7912 linear regulator
*** cap spot C253: Taicon PW, 25V, 100 uF, 6.3x11 mm after 7912 linear regulator
* “startup” capacitor (primary side)
*** Ltec LZG, 35V, 100 uF, 8x13 mm
Caps removed…
And here is the PSU in its recapped form:
For the 5VSB rail, I only changed the bad Ltec LZG, 10V, 2200 uF cap with a Panasonic FM, 6.3V, 2200 uF. The Panny FM is significantly better in terms of ESR and ripple current compared to the Ltec LZG (2470 mA and 0.018 Ohms impedance vs. 1920 mA and 0.030 Ohms impedance.) Deviating from the original PSU cap specs is usually not recommended. But in this case, it’s the 5VSB rail, which often does well even with very low ESR caps like Panasonic FM and FR, along with Rubycon ZLH.
Speaking of Rubycon ZLH, I used two 6.3V, 2200 uF of these to replace the 2x Ltec LZG, 6.3V, 2200 uF on the 3.3V rail. A bit overkill, given that the above ZLH are rated for 1960 mA and 0.020 Ohms impedance (vs. 1525 mA and 0.044 Ohms for the Ltec LZG.) But for a 3.3V rail with a mag-amp design, again very low ESR caps are usually not a problem.
The 5V rail received a different treatment: 2x Rubycon YXJ, 10V, 2200 uF. These are actually rated worse than the 2x Ltec LZG, 10V, 2200 uF caps they replaced in terms of RC and ESR (only 1400 mA and 0.046 Ohms.) Seems that I should have had these and the ZLH caps on the 3.3V rail swapped. But I knew this PSU would go into a modern PC with very little power draw on the 5V rail, so this slight “degradation” in cap specs will not matter. That said, for anyone doing a more than one of these Delta PSUs at the same time, you can just order 5x Rubycon ZLH 6.3V, 2200 uF caps to replace the ones on the 5V, 3.3V, and 5VSB rail with that. This can help with saving on caps when buying in bulk quantities of 10 or more online.
For the 12V rail, I swapped both caps with 2x Panasonic FM, 16V, 1200 uF, even though the United Chemicon KZH, 16V, 1000 uF cap is fine and of good quality. The only reason for that was because I didn’t have any 16V, 1500 uF caps for the Ltec LTG that aren’t motherboard grade ultra-low ESR. So going with the two Panny FM 1200 uF caps gave roughly the same capacitance (2400 uF total compared to 2500 uF before) and similar ESR specs. What I’d like to note here is that the Chemicon KZH 1000 uF cap is 8 mm diameter, but the holes on the PCB are for a cap with 5 mm lead spacing. Thus, using a 10 mm cap is possible – and that’s what I did with the Panny FM caps above. Also, both caps on the 12V rail are connected together without any PI coils or shunt resistors between them, so changing the capacitance of the two caps individually will not cause any imbalance (provided the total capacitance is about the same.)
And finally, for the -12V rail, I only swapped the Taicon PW, 25V, 220 uF cap with a Rubycon YXJ, 25V, 220 uF cap. The other cap filters the output of the 7912 linear regulator for the -12V rail, so it doesn’t really see any ripple or stress. Thus, I left it alone. Also, I tend to trust Taicon a little more compared to the other Taiwanese caps.
Oh, and for the “startup” cap, I don’t remember what I used, but it was likely in the low-ESR category of Nichicon PW and PM or similar. Same voltage and capacitance here, as always.
There are also a few other small caps in this PSU: If I remember correctly, these were 1x 50V, 1uF; 1x 50V, 4.7 uF; and 2x 25-50V (?), 10 uF (I can’t remember exactly, though.) Only mentioning this in case anyone wants to do a full PSU recap. I regularly keep a small stock of 1 uF, 2.2 uF, 4.7 uF, 10 uF, 22 uF, and 47 uF (typically 50V –rated) caps on hand for when I’m recapping PSUs, as these values tend to show up often.
And that completes this DPS-300AB-24b recap.
The PSU is already installed into a PC. I haven’t had to use that PC much, but every time I did, it’s booted reliably, like all of my other recapped PSUs.
Attached Files
Comment