UA 176 Output XFMR - need pri inductance

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

CJ

Well-known member
GDIY Supporter
Joined
Jun 3, 2004
Messages
15,664
Location
California
anybody with an original UA 176 out there who might be able to take some AC current readings at 60 cycles?

some people have complained about low end response on the original, so we are trying to improve on this by lacing the core different (no gap)

knowing where the original transformer primary sits would be helpful as well as interesting!  thanks, cj

to measure, just pull 12BH7a , clip leads to push-pull taps (100%) , take it up to say 10 volts, then measure AC current.

 
CJ said:
anybody with an original UA 176 out there who might be able to take some AC current readings at 60 cycles?

some people have complained about low end response on the original, so we are trying to improve on this by lacing the core different (no gap)
Is there a gap on the original? I would think it was not necessary with a push-pull output...
 
they might have grabbed something off the shelf,

there is a spec floating around somewhere that says 50 Henries at 10 ma unbal DC somewhere, which seems odd for a P-P circuit,
so yeah, it is puzzling.

Putman apparently modified a UTC with taps,

 
Turns out yes original 176 gapped, may be same or similar to 610 (p/n 5946). Later 177 multiple taps not gapped.
Gap definitely not required.
 
I always wonder, is there some minimal gapping in a PP output that makes a good tradeoff against current imbalance and the response loss and distortion that comes with?  Probably not with 12BH7 levels of current. 
 
CJ said:
they might have grabbed something off the shelf,

there is a spec floating around somewhere that says 50 Henries at 10 ma unbal DC somewhere, which seems odd for a P-P circuit,
so yeah, it is puzzling.

Putman apparently modified a UTC with taps,

I've seen the original winding docs for the 5946 transformer, it was not a UTC part.  As David Bock pointed out, it was also used on the old 610 modules.  Use as a push-pull output came first however and that is what it was speced for.  However, it was not gapped despite it being used later as a single-ended plate load in the 610. 
I asked Tom Reichenbach about what I thought was an unusual use (610)  for the 5946 and he thought that the typical manufacturing 'slop' at the time probably meant it was left with enough inductance with the 7mA or so of standing current.





 
So my assumption about gap  was wrong! Proves why an old UTC will work in the 176.
Regarding older xfmrs and DC: for many years I ran a Radio Craftsman P-P power output xfmr single ended with an EL34.  No loss of lows, no meltdown, no heat. As one example.
 
bockaudio said:
Proves why an old UTC will work in the 176.

Even if there's a bit of plate to plate current imbalance, basically, yes 🙂

For a new part, I'll vouch for the Cinemag CM-9661 as a good one for a 176 circuit.
 
i read an old thread where Winston said something like "there were no less than 5 different transformers speced for the 176" but i can't remember which thread, so there could be gapped and ungapped transformers used.

you can gap a core or lace it up to get many different permeabillities  , power transformers are laced 3 x 3 because it is quicker to assemble them this way,  but you can lace an audio core 1 x 1  2 x 2  3 x 3  4 x 4  i have even seen 12 x 12 ,

 
another print of the same thing,  i wound one of these and it had no so great hi end response, since then we rewound the same turns on the same core with much better results , borrowed the winding structure from a Telefunken output coil, split bobbin, reverse wind on the top primary which makes doing the taps kind of weird but no big deal to wind,

note that there is only 3.8 KGauss on the core so there is plenty of Gauss left for unbalanced DC.

i guess we will lace this up 4 x 4 to get some Henries for bass and try it out,
 

Attachments

  • Univerdsal Audio 176 OPT.png
    Univerdsal Audio 176 OPT.png
    47.8 KB · Views: 43
CJ said:
i read an old thread where Winston said something like "there were no less than 5 different transformers speced for the 176" but i can't remember which thread

And, if I said it, I don't remember what I was referring to or in what context.  From what I do remember, most all the transformers were designed by Bill Putnam Snr., it was a part of the design process he kept a hand in and was pretty good at it.  Most were sent out to tender and some would have second or third sources.  I do remember a hand written note regarding the interstage UTC where Putnam suggests to engineering that they look at a different UTC part as an alternative.  I don't believe this particular interstage was ever anything other than an off-the-shelf item though.

C.J.  are we talking about fixing low end deficiency in original units or in "clones"?  Are we sure it's the output transformer at fault?
What's the response with no compression and feeding just the interstage?
 
someone mentioned the low end to me, i do not know if they had an original or clone,

the spreadsheet seems to indicate a potential rolloff problem at the low end based on 50 H.

for an ungapped core the problem goes away as H goes to maybe 2000 and response goes down to 2 Hz.

maybe they wanted to limit the low end to help reduce thumping.

i guess the best way to find out would be to build one,

 

Attachments

  • z1.png
    z1.png
    108.1 KB · Views: 32
OK fair enough.  50H seems really low to me.  I've measured in the 400H+  range for a decent push-pull 30K:600 myself.



 
Thanks!  that was the answer i was looking for!  :D

50 H at 20 Hz = 6.2 K pri reactance    too low for BH7a

400 at 20 = 50 K,    much better for BH7a
 
Old quotes that where captured around here a decade+ ago:

Edit:  I should clarify what I meant by "not UTC's or such".  I mean that they were not an off the shelf UTC transformer.  The input transformer for instance had a Universal Audio Internal part number of IT-5000.  However, the transformer inside the can could have been made by Peerless, UTC, EE or Microtran.  From the documents I've seen, all four brands were used at one time or another.  It depended on who could supply them in a given time frame and at what price per quantity required.

-------------

I do have the full info for the 1008, 175 output (the 5946).  It was made by UTC and, contrary to what I had remembered, also by Electro Engineering Works.
-----

EE and UTC (via TRW/OPT) were both bought by Magnetika.  a few of the 175 and 1008 outputs i've seen have mentioned UTC "pacific division" which i'm guessing could have been a smaller winder bought by UTC (new york based), so who knows if Magnetika even have the design data.  i'm just glad that someone who values the design has a copy.

-----

I'm sure you have this already but, if not and it helps with building 1008's and 175's, the turns for the 5946 output are:

Pri: 30K/7.5K split
Sec: 600/150 split
Tert: 600

Min L with 10mA DC = 55Hy.
 
EmRR said:
Old quotes that where captured around here a decade+ ago:

Wow!  I actually recognize some of those quotes as being mine, maybe they all are?  Anyway, thanks for posting them.  It connected a few more dots in my shot-to-shite memory.


 
CJ said:
Thanks!  that was the answer i was looking for!  :D

OK good.  You're the Coil-Meister General so you'll know how to do it, I don't. 
Just aim as high as practical without incurring other issues.  Forget the gap.
 
Winston O'Boogie said:
Wow!  I actually recognize some of those quotes as being mine, maybe they all are?  Anyway, thanks for posting them.  It connected a few more dots in my shot-to-sh*te memory.

I think they are all yours.
 
EmRR said:
I think they are all yours.

The bit I had doubts about was the info regarding Electro Engineering being part of the sale to Magnetika, I didn't remember ever knowing that stuff.
But it was about 10 years closer to when I would have seen the info firsthand so probably forgivable.

I'm going to go out on a tiny limb regarding the use of the 5946 in the 610 module and speculate that, given how few 610's were built compared to 1008's and 175's, the ungapped 5946 could have been individually selected per module based on whether it passed a minimum inductance spec when run single ended. 
This accounts for why this minimum is specified for when it is to be used in the 610.  There would be no need to specify this minimum if they all met that spec.  And it tallies with Tom Reichenbach's comment regarding manufacturing tolerance slop in the 1960's.





 

Latest posts

Back
Top