[BUILD] CAPI VP2X~500 Series~Preamp Kit~Official Support Thread

[dissonantdissident]

Finished up putting together a few VP25 kits, but I decided to use the Cinemag 3301 for output. I had seen a few people mention using this transformer in 312 builds. Builds went smooth,  but they are lower output, than my other CAPI modules. Basically, they are the same max gain as my others get with the pad switch engaged. Switch still works and pads further. Swapped between the GAR1731s that I built for the project and some Red25s that I keep as my tested opamps, to make sure it's not a build error on the opamp. Pretty much identical output, either way. The units sound good, but super low output.

So, Jeff suggests checking the transformer wiring, between the 2503 and the 3301. Unfortunately, there's no datasheet available for the 3301, on the Cinemag site, but Dave says it is directly swappable with a CMOQ-2 transformer, though.  The CMOQ datasheet is available, so I'm putting it below. The datasheets definitely look different, but I had always thought that the CMOQ and EA transformers all followed the same wire color. If anyone could take a look at them and tell me how to read it or what wires need to be swapped, I'd really appreciate it.

2503 datasheet
http://www.capi-gear.com/catalog/images/gallery/2503/2503-specs.pdf

CMOQ-2 datasheet
http://cinemag.biz/output/PDF/CMOQ-2.pdf

I just put the wires into the circuit, as labeled on the CAPI board. Skipped the T-Pad. Looking at the datasheets, I'm thiunking the problem might be elsewhere. Kinda strange that both modules ended up with the same error. I'm gonna check resistor bands to the BOM, this evening. Any help is much appreciated!

 

[dissonantdissident]

Well, as far as I can tell, transformers are properly installed/leads soldered to correct pads. I'm gonna start checking to make sure I didn't swap a resistor. It would be weird, since both modules are giving the same low output. Totally possible, though! It's always like one little thing.

I've attached a photo of the top of one of the cards. It's a pretty clean build. No solder bridges, underneath. Anybody spots the boo boo, I'd appreciate the help. Kind of stumped, since everything seems to be where it should be going. One of the things I love about CAPI is the ease of building the units.

 

[Potato Cakes]

You it seems that you have the T-Pad installed (the bourns potentiometer is a 600 ohm T-Pad attenuator). Move your blue transformer wire to other solder pad marked Blue.

Thanks!

Paul

 

[dissonantdissident]

Just to make sure that I'm understanding correctly. With the T Pad installed, you can't just solder the wire to the optional pad, to skip it?

I installed the T Pads for cosmetic reasons (ie looking good when in the rack, but thought that wiring into the other pad would still skip it. Turning the T Pad has no noticeable effect on output. I had originally thought that might be the culprit, but since there didn't seem to be an effect from turning the T Pad, I figured that it wasn't interacting with the circuit.

Darn. I was hoping to not have the T Pads on these, but I have no idea how I could possibly unsolder tham.  Guess I should've done more research, before soldering them together.

I'll try moving that wire and see if this gets them operational!

 

[Potato Cakes]

If that doesn't do it, and assuming all solder joints, component placement, orientation, etc are good, then send some tone through it and measure what you're getting at the +/- input of the DOA. If the above statement is true, then it has to be at the input transformer, the DOA, or the output transformer. It's a pretty simple circuit, so finding the problem should be easy.

Let us know.

Thanks!

Paul

 

[Potato Cakes]

After taking a second look, the problem is your output transformer wiring.

Comparing the two data sheet, they have the same colors, but you'll notice that the 2503 has a single set of primaries and three secondaries, while the CMOQ-2 has dual primaries and secondaries. So when you solder the same color leads on your CMQ-2 to the labels on the VP25, you are actually putting one of the sets of primaries where secondaries are supposed to go, so all the electrons are getting very confused, hence, the lower output.

Go look at the schematic on Jeff's website for the VP25. It shows how the transformer should be wired. It is in a 1:2 configuration. Then look at the CMOQ-2 data sheet. Wire your primaries on your Cinemag parallel, with the + polarity going to Red on the PCB, the - side to Brown. The series connection for the output already in the VP25 circuit. The Cinemag green goes to PCB blue, blue to green, violet to yellow, and gray to orange. The PCB violet and gray will be unused.

The T-Pad can stay in place as you had it. That was bad information on my part.

Thanks!

Paul

 

[dissonantdissident]

I really don't understand how to read the transformer schematic, apparently. Just trying to make sure I understand. In the end, I'll need to:

A. Leave Red and Brown where they are, but wire orange to brown pad and yellow to red, as well
B. Swap Blue and Green wires
C. Move Violet to Yellow
D. Move Grey to Orange

Is that right?

Because I'm trying to learn: Looking at the schematic, wouldn't this flip the phase? It would seem that I need to swap the above wires to Red and Brown pads, if blue and green are swapping?

 

[Potato Cakes]

It seems it was late and my eyes were getting crossed looking at the schematic.

So, 50%  yes.

http://www.capi-gear.com/catalog/images/gallery/VP26/VP25-Rev-B-schematic.pdf

The polarity dots on the Cinemag and EA transformers line up per color, and on the VP25 schematic they are all opposite of both of those transformers.

Cinemag Brown and orange to PCB Brown. Red and Yellow to PCB Red. THEN, Green to green, blue to blue, violet to PCB orange and gray to PCB yellow.

Thanks!

Paul

 

[dissonantdissident]

Thanks! Having the outputs and inputs seeming to flip sides really threw me, but these instructions seem right on, and with a little less soldering!

 

[dissonantdissident]

Well, having tried it on one, this doesn't seem to be the culprit. Red and Yellow to PCB RED. Orange and Brown to PCB BRN. Green and Blue stay put.  Violet to PCB ORG and Grey to PCB YEL.

Strangely there's no change, as far as I can tell. Even with these wires changed, it seems that I have the same amount of gain, compared to the unit I haven't rewired and still far below the vp25 that I've already had. Even tried swapping OpAmps, again. Oh well, I'll try and get back to these in the next few days. I had gotten all hopeful, too!

It's kind of weird That the wire swaps seems to have no real difference in output.  Could it be that Red and BRN always went to the 1:2 wired BLU and GRN, making the wire changes unnecessary? Just spitballing, here.

 

[Potato Cakes]

Okay, one more time. I need more sleepy time...

If you look at the schematic versus the 2503 data sheet, you'll noticed that the the secondaries are flipped in relation to polarity, but the connection for the center is still the same. So, look at it being rotated on that center line, then do the same for the Cinemag and line it up with the schematic.

Starting with the top of the secondaries per the schematic.
Cinemag - PCB
Gray - Blue
Violet - Green
Blue - Yellow
Green - Orange

Primaries still as previously discussed.

If that doesn't work then I'll get out my VP25 and do some voltage tests.

Let me know.

Thanks!

Paul

 

[dissonantdissident]

Bizarrely, still no change, even with all the output wires swapped to these different pads. It wold seem like something should make some kind of difference.

Looking at the schematic, I'm even more confused. Then I thought back to the time I made some whistle rock 312s and my 2503-Ls worked perfectly fine, following the color codes marked for the Cinemag (obviously after following the 2503 to 2503-L conversion chart on the schematics. Taking that into account, the more I look at the schematic, the more it looks like your first suggestion was right, Leaving blue/green pair, moving violet and gray over to orange and yellow. Grey and Violet pads are obviously not in the circuit. It seems that this last wire swap would've flipped the phase, but I think that's pretty much it.

 

[Potato Cakes]

Weird. You should be all even Steven. I've used these Cinemags in a similar circuit and it worked like a champ.

One of the things I did notice earlier is you mentioned that the problem preamps had the same output as the other ones with the pad engaged. That seems to me that there is something else going on. I'll pull out one of my VP25's and see if can get some voltages for you to check.

Thanks!

Paul

 

[dissonantdissident]

Thanks a ton. I'll have to order the test jig, but I shouldn't have deleted it from my cart, anyway. This weekend, I'm gonna check all of the resistor bands, too. Caps are all where they're supposed to be.

I'm really just amazed that all of the wire flipping and no audible change as a result. I thought that I had the transformer going out 1:1, which seemed to be a rational explanation, especially with the pad functional. Was also what Jeff recommended looking at.

 

[Potato Cakes]

Yeah, going from 1:1 to 1:2 isn't the same difference as the pad being in/out.

Okay, party time.

Generate a .775 VAC signal at 1k. If you're using a DAW, fire up its test generator and turn it up till the output is approximately -18dBFS. To be sure, take a mulitmeter and put it on pins 2 and 3 and see if you're getting about .775VAC. Some converter boxes are spot on and some aren't, so unless you know that you're I/O box is telling the truth, you need to meter it. Engage the PAD on the preamp before connecting signal to the preamp, with gain all the way down and attenuator all the way up. 

All measurements are with the -/COM side of the multimeter attached to one of the nuts holding the PCB to the L bracket. DOA used was a GAR2520, but they should be similar to a 1731. Set meter for VAC.

DOA In +/-  -  .409 for both
DOA out - .705
Pad - starting top row, closest to the faceplate - .320, .033, .030
Pad - bottom row, closest to faceplate - .320, .033, .030
Middle row resistors (R6-12) starting top left (R6) - 0, .322, .320, .320, .033, .322, .400, .705
Same row, bottom left - R6-R10 - all .030, .402, 0
C10 - both sides - .705
Output transformer
Red - .705
Brown - 0
Orange - .659
Yellow - .049
Green - .049
Blue - .676
Blue Bypass - .676

These are the easiest values you can test with the module fully built. Everything after brown is with a properly wired transformer installed.

Hope this helps.

Thanks!

Paul

 

[dissonantdissident]

Test gig ordered. Thanks for all of the help, I really appreciate it!

 

[Potato Cakes]

No worries. Let us know how you go.

Thanks!

Paul

 

[dissonantdissident]

Well, turns out the mess up was simple. I accidentally printed out the Rev C BOM, instead of Rev B. The main difference is which resistors are listed as R2-4. After I swapped those to the correct positions and wired the transformers per your instructions, these seem to be 100%.

Thanks for all the help!

 

[Potato Cakes]

Excellent!

It seems it is mostly something simple like that. Glad I could help.


Thanks!

Paul

 

[dissonantdissident]

Turns out that the final issue I was dealing with was PSU noise, from my 500 rack. Now that I've got that diagnosed, I'm totally happy with these units. Thanks for the help!

ps- After looking at numerous sites/schematics, it seems that the quadfilar transformers can be wired different ways, but it also makes the schematic confusing. If they can be wired all those different ways that means the second primary wind can be a third secondary wind, like the 2503 schematic shows. I don't really understand how this would work, but if anyone cares to explain it, I'd appreciate the knowledge.

Check out these schematics to understand what I mean about primary moving to secondary:
http://layeredaudio.com/support/index.html