UA 2-610 Troubleshooting

[mitsos]

Hi MAtt,

I completely agree. I don't want to clone this, I want to build the "original" and see what the difference is.  I mostly don't like the reissue, it's unusable (IMO) on things with transients, (sounds too slow to capture them properly. not sure if that's what it is, but it's the only thing that makes sense in my mind.. put it side by side with an API on bass drum). I do like it on some vocals, and some other stuff.

I posted in that thread in the drawing board on inductance, and at least one(respectable) poster's opinion is that the DC through the output trafo would increase 2nd harmonic distortion, that in itself is enough for me to try it.  ;D

Besides, the parts are so generic that if it doesn't work out to my liking, I could rearrange them and have a completely different preamp.

Anyway back to your issue, you get a chance to open that thing up and see?

 

[lassoharp]

Well, I don't know that I'm going into the trafo-winding business

I didn't think you were.  Was just kind of expressing some enthusiasm for the ambitious attempt.  I hope you nail it!


I can't speak much for the reissue 610 as I've only heard it once as a demo model.  A good low ratio SE OT could certainly be used for beefier output stages that should pose no headroom problems for DAW inputs.

There is a thread on the earlier 100D console amps that have roughly similar output stage.  I haven't seen the full console schematic but I assume those amps were feeding a passive mixing bus and there was a hefty program amp to drive the recorder.  The 610/headroom topic has been subject of much debate.  Hard to tell what's going on there.

 

[mitsos]

I've seen the 100D but the trafo has a tertiary winding which is part of the feedback circuit, and I've avoided it for the extra complexity. But that's the circuit this thing will become if I end up not liking it.

The truth is I really like transformers.

 

[rickc]

I just bought a UA LA-610 MK-II. It's the combo of a 610 preamp, a 2-knob equalizer, and an LA-2.

I've already popped the top and looked inside.  It's a combination of thru-hole and SMT. There are a few ICs on the board, one I suspect is the meter driver, and another is the driver for the dc-dc converter.

Let's start there. The power supply is one of those OEM things that delivers 24v DC. No large hum field to deal with either.
there's a DC-DC converter on the single PCB that turns the 24v into 300v for the B+ supply. Apparently, this converter also makes P48. Maybe it makes heater voltage for the EL84.

Since the only power on the board is 24v, it must mean that the tube heaters are wired in series; there aren't any resistors that are big enough to be dropping resistors.
There's an EL84 on the board that drives the EL panel. In the vintage units, that's a 6AQ5, which  is a 6V6 in a 7-pin jar. I don't know how the EL84 gets its filament power. That bears some investigation.

There is no B+ on the output transformer. The transformer is a PCB mount device made by CineMag, and I probed all of the transformer pins looking for voltage. I found none.

They use relays for nearly all switching. Although this adds cost, it also puts the switching at "point of usage" which is a great idea.

The EL cell is a small circuit board, with the EL panel mounted on one side of a PCB, and two photocells mounted to the other side. It mounts on the main pcb via two header connectors. A black plastic box screws down onto the circuit board to make a light-tight enclosure. The photocells are in transistor-style metal cans, but larger than a TO-5.

Now to circuit observations.  It appears that every stage is cathode biased, and there are no cathode followers. That says that the output stage for the compressor section is not the same as in an LA-2. All of the cathodes measured some smallish voltage above ground 1-4 volts.

Tubes are:

ECC83/12AX7 for the first stage of the mic preamp
ECC82/12AT7 for the 2nd stage of the mic preamp (line driver)

ECC83/12AX7 for the Compressor
ECC81/12AT7 for the compressor
EL84/6BQ5 drives the EL cell

the duo-triode tubes are all made by groove tubes.

After looking at the one schematic I could find posted here, I was curious about the effects of the gain switching.

It looks like for the plus gain steps, the gain of the preamp is just increased, as the clip point decreases with increasing gain, as I'd expect it to, but for the minus gain steps, there is a decrease in output level (determined by leaving the LEVEL control at a fixed setting that did not exhibit clipping) and the increase in signal needed to cause clipping did not appear to track this. I will make more exacting measurements later.

The pots look like common-ordinary carbon track units. They are not hot-molded composition controls. The switches in the EQ section look like Grayhills, but the other switches that run the input section, gain switching, etc. are open-frame units of no particular provenance (some brand of asian). Since they use relays for most switching functions, I believe the quality of these switches is unimportant as far as the audio is concerned. The VU meter does not look legit (it looks like a microammeter with a VU scale), some tone bursts will reveal the truth.

Clipping, when it does occur, is asymetrical.

That's my report, for now.

 

[mitsos]

thanks for the breakdown, seems they've changed it from the first version, at least according to what is written here:
http://mixonline.com/mag/audio_universal_audio_la/

I didn't have a chance to look at the 2-610, I'll try to do that this week, but I'm extremely busy these days so this stuff is taking a back seat. I'm guessing it will be more or less along the same lines as yours anyway.

I've heard from John H about the trafo, he doesn't have the info but gave me a lead on someone who might still have one, I'll try to get some info, but from his comments it doesn't seem to have much of a secret, it's just a simple output possibly meant for something else and Putnam used it in the 610 because they had already made them for the 175 comp.  Otherwise, CJ posted way back that a certain RCA output would be sufficiently close, though it states 50/50 nickel, and I can't see that being good with DC on the primary, but I have the lams, so I'll give it a shot and see what happens.

I'm nowhere near building the rest of the circuit of course, one thing at a time!

 

[lassoharp]

certain RCA output would be sufficiently close

If what I think, you will not have the high freq response that one would be accustomed to with 4:1.  Headroom as well.  How much current will your version be running at that stage?

 

[mitsos]

HF response? Who said I wanted that anyway?  ;D  kidding of course.

According to the schem, 7mA, but JHinson said it should be more around 5mA.

for reference, here is the thread where CJ posted about the RCA (a few posts down):

http://www.groupdiy.com/index.php?topic=37944.0

This I can build, I have some nickel EI625 lams as well as M6, so I could mix and match to get some of the "DC capabilities" of the M6 and the inductance of the nickel.  It's a higher ratio than I want (10:1, vs 7:1 in the UA), so I won't be using it "as-is" but I figured I could re-calculate the primary turns to give me as much inductance as possible. In any case, it is all experimental.

People have used the cinemag CM9661 for UA varimus which used to use the 5946 (Ed once suggested it and the CM30K/600 to me for this), I think one is gapped, but both can be ordered that way, which I don't want to do initially. Other than that, I wish they showed a mechanical drawing, too lazy to call internationally right now and ask.


There is also some interesting talk here:
http://www.groupdiy.com/index.php?topic=2580&start=0


sadly, found out that Jhaible passed away recently.

 

[rickc]

the output transformer looks like the one shown in the company's website. But up close, you can see that it is an EI lam stack, with a single end-bell on the top. That's marked CineMag, but I don't recall seeing a number. The bottom side is a pcb header for insertion into the board, so there's no bottom end bell that I could see. There are many pins on the header, but I don't know if they are all used for connections (vs something purely mechanical)

 

[mitsos]

Hi Rick,

you're talking about the LA610 output trafo, right?  I'm sure it's different than the one in the original 610, that was John Hinson's point in various posts. We'll see what I come up with, I talked to my winder buddy, he's got #37 and #40 wire, and I have the lams, so hopefully I can get something done. I'm even going to try to do it with cardboard instead of a bobbin, just for kicks.

 

[rickc]

mitsos: yes, I'm talking about the LA610 output transformer.

the dc-dc converter on the main board also makes 5v (!) for the tube filaments. all measure the same, so they're running in parallel.
that also means that the same converter makes the p48.
It appears that the converter runs at about 416 khz.

 

[mitsos]

I wonder why they use a DC-DC converter for the LA610 and not for the 6176? Or maybe they're still working on the 6176 Mk2?  I've seen the inside of the older LA610 and off the top of my head it had grayhill switches, and  think a linear PSU, but I don't remember for sure.  I'm not a PSU snob, in fact, that's one thing I would love to see here, a DC-DC converter designed from the ground up for audio.  Igor once mentioned something, and when I saw his "universal switcher" I got excited thinking, wow, someone did it, but alas, it was "only" a PIC/relay switching circuit. anyway, that's all I've got here, I'm gonna get my thoughts together and start a post in the drawing board to see where we get with this trafo.
thanks for the input!

 

[rickc]

They used a 'canned' supply between the mains and their board. this was a fast and easy solution, and perhaps their run numbers aren't big enough to justify doing their own. Besides, the biggest reason for using a canned supply was that it gave them instant UL approval for that supply, it let them use a universal supply, and perhaps most of all is that it saved them a ton of money by not having to go the full-blown approval route. It's the same logic for using a wall-wart, and it has the distinct advantage of living inside the chassis. Huge win for them. Agency approval is expensive.

so, even for us DIY types, I think that taking a lesson from this is prudent, as our home-made products have no UL or other agency approval, so the best we can do is to follow accepted practice. But if something untoward could happen (and you know, Mr. Murphy has a passkey to ALL of our homes), insurance weasels could blame it on our unapproved gear, which might be a huge PITA.

In my own gear, I make sure that everything tied to the AC mains is UL recognized, down to and including the wire.

 

[mitsos]

Easy UL listing is the main reason I can think of for them going that route. What is surprising to me is that they didn't do the same with the 6176 (610 + 1176 "strip").  It would be cool to find an "affordable" UL-listed supply such as this, that could be bought in lowish numbers. 

 

[rickc]

I think any of the 'canned' linear or switching supplies sold by Mouser or Digikey would carry the UL mark.

Although I can't imagine why they didn't go this route before, but maybe they finally finished the design of the on-board switcher, which let them move ahead. Now one thing I don't know is whether they needed to get FCC type approval on the box.

Time to take the $3 spectrum analyzer down to the bench (am transistor radio).

 

[lassoharp]

insurance weasels could blame it on our unapproved gear

"thou shalt not utter into existence in any form the ammo for loopholes to said weasels"

DIY code article 37.    ;D

 

[rickc]

Lassoharp: rat on!

 

[lassoharp]

If someone needs a specimen. (Not mine)

http://www.ebay.com/itm/PAIR-VINTAGE-UA-UNIVERSAL-AUDIO-1008-TUBE-MIC-MICROPHONE-PRE-AMPS-AMPLIFIERS-/320840030009?pt=Vintage_Electronics_R2&hash=item4ab38e5739#ht_500wt_715

 

[rickc]

I have seen this seller before, and I feel he has an over-inflated belief in what this stuff is worth. He does come thru if you win the auction, but I think his price is delusional, to say the least.

 

[lassoharp]

I have seen this seller before, and I feel he has an over-inflated belief in what this stuff is worth. He does come thru if you win the auction, but I think his price is delusional, to say the least.

Very much agreed.  His copy is . . . 'thick'  too, and maybe a bit misleading on many items.

 

[mitsos]

Hey guys,

So, as I was planning to start a thread on winding the output trafo for the original 610 (and 175,176,177) when I got a message from one member who led me to another member (don't know if they want to be named, but maybe they'll speak up here), who led me to Cinemag, armed with a part number: CM2092CF (deleted wrong info) I was told they could make it in steel or nickel, butt stacked (AKA gapped) or not. 
That's all the info I have so far, I don't know if the primaries have a center tap or all the other taps (for the 177 for example) but it only costs $27 in steel so it can't be that complex of a trafo.  I'm going to pick up a couple to try out, no idea when I'll have them cuz where I live I have to deal with month long customs delays ....    :


EDIT: talked to David Geren just now, the 2092 was a prototype made for someone (didn't say which company) as a replacement for a UTC PA6003, not the 5946.  I don't know what the 6003 is, but david's notes said something about a 12AU7 stage.
so back to the drawing board I guess for the PA5946, although I may end up trying this "little guy". It's the size of an API 2623.