[BUILD] CAPI Dual VPR & 51x Floor Box PSU~Official Support Thread

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Hi toolboxmuziq,

I'm not experienced enough with the build to really say, but the thing that immediately comes to mind is that there are five regulators attached to the sides of the case. They all look the same, but one of them is different, which is something that I know I almost missed myself when assembling it. Might be worth checking...

Kaz
 
Yeah , be sure they aren't over tightened , have the insulators and not making contact with ground , I believe the build
instructions tells how to check for that
 
I think I've blown the regulator for -v#2 so I'll have to order and replace it. After removing and replacing the regulator at +16 I still am not get getting anything. I've tested at the place where the cord should be installed but I get no reading. Someone please help.
 
toolboxmuziq said:
jsteiger said:
Blown fuse? Probe on both ends of the fuse holder(s).
I checked both ends and I get nothing from either! Can anyone suggest the next step?

Again, I'm no expert, but I would do the following:

1. Make sure the PSU is unplugged to avoid accidentally causing more issues with accidental shorts.
2. Use your continuity tester.
3. Look at the schematics posted in the second post of this thread.
4. Notice that a lack of continuity between the outputs at the molex connectors and the fuse holders is pretty much impossible!

Like I said I'm not an expert and I was probably fairly lucky with my build, but if I was in a similar situation to you, I would keep one probe on the output terminal, and move the other upstream a component at a time by looking at the schematic and confirming continuity. A better approach would be to move both probes upstream a component at a time while maintaining the offset, so that you're probing between the output leg of the upstream component and the input leg of the downstream component. That would confirm that your circuit is connected, but wouldn't rule out component failures. Only when I'd confirmed continuity and checked every part against the schematic would I try switching it on again, but that's just me.

Would this be the right approach? I don't know...I'm a software guy, but that's where I would start. For larger circuits I would modify that approach to sub-circuits (testing for expected voltages), and use a binary search technique, but I don't think this circuit is complicated enough to warrant that...

Kaz

 
Can someone tell me what should be the appropriate readings coming directly from the gdiy 51x toroidal transformer and the screw terminals. I have traced and confirmed that I do have a power flow but it's only reading .300 at the +16 terminal and throughout the +16 section.
 
Hello,

did you measure directly at the wires/screwterminal that is coming from the torroidial transformer,so meaning the secondaries?
You must set your dmm to ac voltage,earliest stage for dc readings is after the rectifier!

Let us know,

best regards,

Udo.
 
Folks, i'd like to bring to your attention something.    Commercial racks do not allow for over-clocking the PSU, and as such commercial modules may or may not have protection diodes, which you need the extra 0.7V.    As such, if you're running commercial modules, it's smarter to run your PSUs right at +/- 16V than at +/-16.7V.  Commercial modules will work fine when underpowered (15.3V if they have protection diodes), but it's unknown how they'll respond when overpowered.    Radial, API, Purple boxes: you can't overclock their PSUs.  If you use any commercial modules in your rack, they're expecting at most +/-16.1V.  A commercial developer told me this.
 
Interesting thoughts Chuck. I have not personally seen all available modules for the 500 series format. I am surprised that manufacturers/designers would not include protection diodes. I know for a fact that API has them in every module they produce. APSI also has them. I don't recall seeing any kits around here without them but I have never looked specifically just for that either. My guess would be that most commercial 500 series products have them.
 
Hey guys...

As someone who has recently built one of these PSUs, and previously posted about having successfully biased the voltages as specified in Jeff's guide, I feel like I want to come back and contribute something, now that I've had time to live with it a little bit.

FWIW, I'm a DIY newbie and new to 500 series too, so take everything I say with a pinch of salt!

I've got a couple of pre-amps (two of the same running off the 16v rails) and a couple of compressors (different, one running off 24v rails, one running off 16v rails) running in my rack from the usual sources, and during the setup and config of one of the modules, the numbers I was getting out of the probes weren't matching the expected numbers mentioned in the setup guide itself. While not detrimental so far, I went ahead and queried the creator of the module on their stance regarding voltages. I won't say who I asked to avoid putting words in their mouth, but it is clear that their modules expect something closer to the standard +/-16v of the rails rather than the +/-16.7v I've currently got them biased to.

Having given this some thought, I think it's fair to say that a specification is only useful if it is adhered to, and Jeff's build document clearly shows the voltages on the rails as specified as +/-16v, +/-24v, and 48v in the two (500/51x) connector pinout specification diagrams.

It is incorrect to make assumptions about what is happening on each individual module after that, as any module designer, whether commercial or not, has to work assuming the spec is being adhered to and with that, is free to make trade-offs as they see fit to produce a useful product. One such trade-off might be having protection diodes if the full +/-16v isn't required, or might be omitting them if they need every little last volt available to them.

So, someone who works to the spec, but decides to use protection diodes, must surely be doing the following calc when designing the rest of the circuit:

16.0 - 0.7 = 15.3

That means that 15.3 is what they built the rest of their circuit to work with. Giving them the extra 0.7v to work with doesn't do them any favours if they aren't expecting it, and actually might decrease the lifetime of some components if they are operating near their limits.

If on the other hand the designer omits the protection diodes because they want the full +/-16v to be available to them, then still, that +/-16v is what they're expecting to drive the rest of their circuit, not +/-16.7v.

Logically, the only people designing their modules to handle the extra 0.7v gracefully are the people designing just for 51x racks and with specific knowledge of the biasing that is de-facto standard, but as soon as they do that, then they're not actually working within the api500 spec as non 51x racks don't have the ability to drive the +/-16v rails at higher voltages.

Of course, I can't comment on what the other rails should really be, but my instinct is that any implementation of a specification must match the specification as closely as is reasonable possible.

It would be nice if the various alliance members could come forward and put forth their final word on this, as I'm just working off my own sense of logic here, but I'm now starting to think that it's right that we try to trim our PSUs to hit the specifications as closely as possible. I plan to do this myself when I get a moment, and intend to to be within +0.1 of the target voltages.

Anyway, it would be great if anyone else with an opinion on this stuff could chime in, and also anyone who has a commercial 500 series rack, as it would be nice to find out a little bit more about the range of voltages that are being typically supplied.

Oh, and I just want to say sorry and apologies if this comes across a little harsh or ungrateful. That's not the case, I really appreciate all of the hard work that everyone has put in to making this stuff available for all of us average joes who wouldn't be able to do it by themselves! ;)

Just trying to get to the bottom of this, and think this issue needs a little clarity...

Kaz

 
Most times it doesn't matter, pointless to discuss.

We are talking about analog audio circuits ...

If there is any exact voltage reference needed, it is (or should be) in the circuit.

So it shouldn't matter if you have 15, 16 or 17 volts.
Most OpAmps are rated up to 18V.
PRR would mention the +/-10% mains voltage etc.

If a module doesn't work within those variations, it is a bad design.
 
[silent:arts] said:
...

If there is any exact voltage reference needed, it is (or should be) in the circuit.

...

Hi Volker, thanks for chiming in!

So I think what you're saying there is that it would be bad practice to use the voltage on the rack rails as an input where an exact voltage is required, even if theoretically it was the exact voltage you needed? (let's say an imaginary digital component that required 16v...speaking completely hypothetically here)

Is there some reason why that's bad practice? Is it because all of the other modules in the rack mean there are more unknowns that can't be accounted for anyway, so some kind of tolerance has to be built in to a module in the first place? Or is it because there are too many different rack implementations with the associated variances to be able to blindly rely on it being accurate?

I'm only a software guy, so not trying to be a dick, just trying to understand the issues (one day I hope to be able to contribute with more than just questions!). ;)

Oh, and regarding the +/- 10% mains voltage...shouldn't that be a non-issue because of your lovely psu pcb? Surely that +/- 10% mains voltage variation never impacts on the voltages that reach the rails, except during brown-outs? (otherwise, you know, what's the point of those regulators and capacitors?)

Thanks for your patience!  :D

Kaz

 
What I'm saying is it doesn't (shouldn't) matter if you have 15 or 17V.
If it does it is a problem in the module.

Regarding the PSU PCB we sell: it is not digital ;-)
It is a percentage from the incoming voltage.

So, in Germany where we should have 230V Mains, once the "Tagesschau" begins, it will be lower.
 
[silent:arts] said:
Regarding the PSU PCB we sell: it is not digital ;-)
It is a percentage from the incoming voltage.
Ok, so I think this is the important bit for me. It certainly spells out the need for being resilient to < 16v voltages. I need to read up on those regulators to understand how > 16v can happen (transiently) though. I was under the impression that those voltage regulators impose upper limits quite well.

[silent:arts] said:
So, in Germany where we should have 230V Mains, once the "Tagesschau" begins, it will be lower.

I had to google "Tagesschau" to understand that one... ;) For everyone else: http://en.wikipedia.org/wiki/Tagesschau_(Germany)...

Kaz
 
Kaz, all fine.

set it to somewhere between 16V and 16.7V and you are done.
nothing magic in here, and the discussion is as useful like splitting hair.

the regulators will keep the adjusted voltage, even with the Tagesschau and Oktoberfest!
(I'm NOT a fan of the last one. Lived too many years next to it)
 
[silent:arts] said:
the regulators will keep the adjusted voltage, even with the Tagesschau and Oktoberfest!
(I'm NOT a fan of the last one. Lived too many years next to it)
Same here,don´t like it....but it´s still a heavy current-sucker ;D

Greetings,

Udo.
 
Thanks for the discussion guys, it's certainly helped me think about the various issues.

[silent:arts] said:
set it to somewhere between 16V and 16.7V and you are done.

That's the plan, though I'm not in a rush to change anything particularly, as I don't feel like opening up the PSU right now... (will probably get around to it sometime next week).

In other news...got a mouser shipment arriving today, 2xEQ to finish off and 2xpreamps to start and finish! Yay!

Suppose I'd better get back to my day job...luckily my boss (me) hasn't noticed that I'm not working on my game right now... ;)

Kaz

 
Back
Top