The new Zeners did the trick. The music is playing and all measurements are within tolerances. Thank you so much for all your help. This has been a lot of fun. And I've actually learned a few things in the bargain!
J
I have been searching this thread (whew, it's a long one!) looking for info on stacking these boards. I want to stack two fully populated boards for 120k uF per a rail. My question is, how do I connect the vias for this? I'd like them to be in parallel and want to do it right rather than chase my tail.
I've stuffed two with 15k uF filters, 4x 0.47 resistors, 4.7k bleeders, plus the LED. I've ignored the snubbing. Do I need to connect side to side grounding on the top board? I'd guess yes, plus the bottom to top jumpers, give me a parallel cap bank like I expect with no ground loops or magic smoke.
This will be running an F5Tv2. Originally I was using a Chinese diy board (yeah I know, not the best idea) but it was noisy and wouldn't stop. There was no great spot to star ground either, a feature I look forward to on these main boards. It's a great sounding amp once it's over the noise floor, I'd love to get it working with no hum!
Thanks in advance.
I've stuffed two with 15k uF filters, 4x 0.47 resistors, 4.7k bleeders, plus the LED. I've ignored the snubbing. Do I need to connect side to side grounding on the top board? I'd guess yes, plus the bottom to top jumpers, give me a parallel cap bank like I expect with no ground loops or magic smoke.
This will be running an F5Tv2. Originally I was using a Chinese diy board (yeah I know, not the best idea) but it was noisy and wouldn't stop. There was no great spot to star ground either, a feature I look forward to on these main boards. It's a great sounding amp once it's over the noise floor, I'd love to get it working with no hum!
Thanks in advance.
Can you sketch up a quick schematic of what you're trying to accomplish? Perhaps start with the the schematic for the board and modify it to illustrate. Maybe a bit of the "why" behind it would be useful also.
I may be misinterpreting what you're wanting, but if you only want more capacitance per rail, you would not "fully populate" both boards. Also, the layout for those boards is for a CRC filter. So, it would be important to know where you want the extra capacitance. First C, second C, or evenly spread? What I think you're suggesting may be possible, but I'm not sure it's practical. I'm also making the assumption that parts shortages / prices kept you from simply just buying individual caps with more capacitance. So, suggesting that you just buy 30kuF caps won't do you much good, right?
Is one of these "stacked" PSUs going to be for one channel (dual mono or monoblocks) or are you intending to use a single "stacked" one for both channels?
I may be misinterpreting what you're wanting, but if you only want more capacitance per rail, you would not "fully populate" both boards. Also, the layout for those boards is for a CRC filter. So, it would be important to know where you want the extra capacitance. First C, second C, or evenly spread? What I think you're suggesting may be possible, but I'm not sure it's practical. I'm also making the assumption that parts shortages / prices kept you from simply just buying individual caps with more capacitance. So, suggesting that you just buy 30kuF caps won't do you much good, right?
Is one of these "stacked" PSUs going to be for one channel (dual mono or monoblocks) or are you intending to use a single "stacked" one for both channels?
Here is my cruddy drawing and an explanation.
Basically I want to combine the capacitance of two boards, as a single board will not power the F5Tv2 without very large (and expensive) capacitors.
If this config is not possible, I will have to order larger caps and try again. I can always use the other caps in a different project.
Also I would like the power evenly spread. I don't see a reason to put it on one side here. This is not a dual mono. I intend to use one power supply for both channels.
I can't find 30k uF snap caps that fit. They're all screw type and I don't want to deal with that really.
Also, the diyaudio store said this was possible, or so I took it. I could be quite wrong and they meant stack for dual mono.
From what I understand, it wouldn't be good to power both boards from a single pair of rectifiers, but I could be wrong. I'm still learning, only been at this a year with no EE training.
I can't find 30k uF snap caps that fit. They're all screw type and I don't want to deal with that really.
Also, the diyaudio store said this was possible, or so I took it. I could be quite wrong and they meant stack for dual mono.
From what I understand, it wouldn't be good to power both boards from a single pair of rectifiers, but I could be wrong. I'm still learning, only been at this a year with no EE training.
Last edited:
You didn't answer the question re: monoblocks / dual mono or standard stereo build.
Were I in your shoes.... I would use each board that you already have, with the caps and parts you already have and build it dual mono. Is that practical for you? The example PSU that Nelson Pass presents in the F5 Turbo article is 40kuF => R => 40kuF for each rail. (page 11)
https://firstwatt.com/pdf/art_f5_turbo.pdf
You'll have 30kuF => R => 30kuF for each rail on each board. If you go dual mono, that (IMO) is potentially better and would be sufficient. However, another opinion or two would not hurt.
If you can't do that, then it does look like the boards can be "stacked" for a parallel cap bank if you want.
See schematic attached. Tie V+ to V+; V- to V-; GND to GND; D+ to D+; D- to D-; etc. etc. What I am unsure of is based on the board layout where exactly the R would fall in the ultimate CRC configuration once paralleled. Someone that knows this better than I would need to help you out re: whether it's most advantageous to populate the resistors on one board or the other or on both boards and/or if any other jumpers would be needed.
Rather than cause confusion, I'll wait for someone else to provide the definitive answer re: the CRC and maybe chime in on the overall idea. I'm sure there's someone out there that's tried it.
Good luck.
Were I in your shoes.... I would use each board that you already have, with the caps and parts you already have and build it dual mono. Is that practical for you? The example PSU that Nelson Pass presents in the F5 Turbo article is 40kuF => R => 40kuF for each rail. (page 11)
https://firstwatt.com/pdf/art_f5_turbo.pdf
You'll have 30kuF => R => 30kuF for each rail on each board. If you go dual mono, that (IMO) is potentially better and would be sufficient. However, another opinion or two would not hurt.
If you can't do that, then it does look like the boards can be "stacked" for a parallel cap bank if you want.
See schematic attached. Tie V+ to V+; V- to V-; GND to GND; D+ to D+; D- to D-; etc. etc. What I am unsure of is based on the board layout where exactly the R would fall in the ultimate CRC configuration once paralleled. Someone that knows this better than I would need to help you out re: whether it's most advantageous to populate the resistors on one board or the other or on both boards and/or if any other jumpers would be needed.
Rather than cause confusion, I'll wait for someone else to provide the definitive answer re: the CRC and maybe chime in on the overall idea. I'm sure there's someone out there that's tried it.
Good luck.
It's a standard build. Dual mono would fit if I bought two new transformers (trying to avoid that) and drill a few new holes to mount them.
Dual mono would be ideal, but I did not plan my first build well.
I had the same issue reading the schematic. It does seem your soldering advice is correct. The resistors also confuse me a bit. They may just need a jumper, but I'd think the resistors would be needed for the second board, with this schematic.
Thank you for your time and good wishes.
Hi guys,
I'm planning to build an external power supply for my future FW and ZM amplifiers. From what I understand, an external PSU requires the installation of an extra capacitor bank in the amplifier chassis.
Since I'm not absolutely sure how to make the wiring/electrical connections between the two 'sections' of the PSU, I have made a sketch (depicting only one channel) that shows my blueprint for these connections.
I would appreciate any corrections or comments to my sketch.
My external PSU will have 120000uF per channel (CRC - 60mF/0.1Ω/60mF). Will 20000uF per channel be enough in the extra capacitor bank inside the amplifier chassis (RC - 0.1Ω/20mF)?
Or, should I go higher? 40mF? 60mF?
As a disclaimer note I have to say that my sketch was based on Octopus Technicus brilliant diagram for Aleph J's PSU.
Thanks.
I'm planning to build an external power supply for my future FW and ZM amplifiers. From what I understand, an external PSU requires the installation of an extra capacitor bank in the amplifier chassis.
Since I'm not absolutely sure how to make the wiring/electrical connections between the two 'sections' of the PSU, I have made a sketch (depicting only one channel) that shows my blueprint for these connections.
I would appreciate any corrections or comments to my sketch.
My external PSU will have 120000uF per channel (CRC - 60mF/0.1Ω/60mF). Will 20000uF per channel be enough in the extra capacitor bank inside the amplifier chassis (RC - 0.1Ω/20mF)?
Or, should I go higher? 40mF? 60mF?
As a disclaimer note I have to say that my sketch was based on Octopus Technicus brilliant diagram for Aleph J's PSU.
Thanks.
One of my goals for having an external power supply is to use it with multiple amplifiers which saves money on transformers and capacitors. If I were to put the PSU filter board in the main chassis, I would be largely defeating that purpose since I would have to buy all the PSU capacitors for each of the amplifiers.
I am no expert, so I cannot give you a proper technical explanation, but we have extremely knowledgeable members who can explain this to you.
However, I think it is intended to eliminate any noise picked up by the umbilical, as well as to achieve low inductance and resistance supplies.
Furthermore, it seems to be generally agreed that the power transistors should be as close to the reservoirs as possible.
I am no expert, so I cannot give you a proper technical explanation, but we have extremely knowledgeable members who can explain this to you.
However, I think it is intended to eliminate any noise picked up by the umbilical, as well as to achieve low inductance and resistance supplies.
Furthermore, it seems to be generally agreed that the power transistors should be as close to the reservoirs as possible.
Last edited:
@Analogico if it were me, I'd add a 4th wire to the umbilical for chassis ground and tie the two chassis directly together (ie: no ground-lift back to the capacitor bank in the amp chassis).
Also, there's no need for 4 caps in the amp; 2 will do (assuming you can get enough uF with 2). I have no idea how much uF you need.
Also, there's no need for 4 caps in the amp; 2 will do (assuming you can get enough uF with 2). I have no idea how much uF you need.
Most important specs are:
1. VDC rating > PSU rails
2. Diameter <= 35mm
3. Pin # and spacing. 10mm "snap-in"
4. min 10kuF per cap. anything above 15kuF is overkill
The BOM is a good guide.
Quick look at the ones you posted, way overkill but should work.
Essentially yes, they would be OK. Maybe a bit low voltage, by the time it's rectified that would produce around 30v rails.
You would absolutely need to use some form of soft start though. The inrush current would be quite something.
With 30v rails and a 500va transformer I would assume your probably building some varient of class a amp. In which case there's only so much much the transistors are going to be able to sink so maybe 47000 is a bit much. Have a look at esr. If 15000 or 22000 caps have significantly lower esr they may be a better bet.
You'll also want to check the ripple current rating of the caps. I can't remember exactly what to look for, but maybe search the forum for more info.
I completed and tested the diy universal power supply in february and it tested fine (dual mono blocks). Shelved it until last month while waiting for the Aleph J boards to become avaialble. Built the Aleph J boards but having issue with one of them. Want to rule out the PSU. For the life of me I can't find test procedure for the PSU. I am a total non electronics person. The one I used earlier was very specific, ie. put the red probe here and the black one here and you should get this voltage. I've not hit on the search term that gets me what I need. Hoping someone here knows exactly where it is. May have had a short video showing the dim bulb tester in operation. I know it's in a really obvious place.
This blog may help, specifically the PSU section.
https://diyalephj.blogspot.com/
Edited for better link to intro page.
https://diyalephj.blogspot.com/
Edited for better link to intro page.