ADM Transformer 3-010-5197-0 Take Down

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
we had some trouble with the Fish and Game Dept a few weeks back, seems they don't like me  buying tortoise shell guitar picks made in Japan for twenty bucks a pop,  (there is nothing quite like a tortoise shell pick if you have never used one, the wetter they get, the tighter the grip, and they have a certain hi end that is hard to describe, and they last forever)

after we refused delivery of the illegal pick for obvious reasons the  mailman was arrested and is now an ex felon, finding it hard to  get a Federal work, 
the choppers went away, but a few weeks later here comes this ADM transformer from rackmonkey and i'm like "here we go again".


 

Attachments

  • tasmanian.JPG
    tasmanian.JPG
    129.8 KB · Views: 41
here is what we have so far,

does anybody have the circuit? i wanna see if i gots it backwards,

looks like it was mean to be driven by a strong transistor if 11 Henries is all we have,

very flat which is expected by low ratio and low DCR,

we will get some sq wave shots and do some inductance plots,  and then it will be 1st degree murder,  :D
 

Attachments

  • ADM Output.png
    ADM Output.png
    23.1 KB · Views: 60
pri ind vs level

Henries came up from initial measurement at 2 volts,

could because the xfmr has been unplugged for eons, or we had some magnetization from taking DCR readings,

we will find out after we get the wire off the core. nickel  will magnetize then d-Gauss with hi input level at low freqs,
steel domains will loosen up  after injected to hi levels,  which will raise H

some freaks leave new OPT's connected to hi level 60 CPS overnight to break in core,

Ni will also de-Gauss with hi levels,

pri sat at 8 V-ac, seems low,  must be Ni core,
 

Attachments

  • adm opt1.png
    adm opt1.png
    8.3 KB · Views: 24
here is Inductance vs Frequency,


looks like a tape wound core if it is a torroid, probably 50% Ni


powder core would have been flatter at the low end,


gapped C core would have also been flatter.


i bet it is National Arnold with green epoxy,


probably 1000 turns primary 500 secondary









 
very good overshoot test,

top pic at 1 KC  has injected signal superimposed by output, sometimes sq wave output signals will distort input wave form of generator but not here.

 

Attachments

  • adm sq wave.JPG
    adm sq wave.JPG
    179.1 KB · Views: 21
You good fella, CJ! Thanks for doing this (yet again)!

These are the ADM outputs you see on so many of their modules from the 70s and 80s. I picked up a lot of 5 off Reverb for $50 for the lot. Model no. 3-010-5197-0. Check the pics of any 70s or early 80s ADM module and you're likely to see this one on the output. Thought it'd be good to have it documented.

Tom Chrapkiewicz was evidently an engineer at ADM in those days. He commented on a post on GS about these transformers:

"The output transformers were provided by several small custom transformer houses in the Detroit - Chicago area. These 'houses' were often power transformer suppliers, but we assisted them in material selection and stringent audio specifications to ensure adequate audio performance. [This was in the era of HP332/339 Audio Analyzers.] One of the suppliers was 'SHC' as I recall."

'TomC'
Thomas Chrapkiewicz


2 of the 5 transformers in the lot I picked up were the model CJ is tearing down here, which is the model I've seen the most (mostly in pictures). The other 3 (which I originally thought would be the same, just without the model number printed on top) are actually different. Same can, but you can see the toroid through the clear epoxy on bottom. Different DCR and inductance measurements. Here's what I measured on those:

Same pinout as the one CJ has here:

2-pin side: 16.89 ohms DCR, 3.23H @ 100hz (serial) and 4.35H @ 100Hz (parallel)

3-pin side: 127 ohms DCR, 16.37H @ 100Hz (serial) and 20.24H @ 100Hz (parallel) (center pin is shield)

measured using a DER-EE DE-5000 LCR meter.


Pictures:

AZFRKw1l.jpg


nxtwdSJl.jpg


No idea when one model was used vs the other.

Interesting comments on this post at Preservation Sound from a slew of ex-ADM guys and our resident ADM expert Brian Roth. A lot of historical nuggets here:

http://www.preservationsound.com/2012/09/adm-consoles-of-the-1970-and-1980s/
 
CJ said:
...does anybody have the circuit? i wanna see if i gots it backwards...

Here's a 2780 input module schematic. I have a picture that shows that it uses the same transformer (based on part number).  2N3053/2N4037 complimentary pair if me eyes are seeing right (kind of a blurry scan), so your theory about a beefy output stage was correct.

620 ohm terminating resistor on output.
 

Attachments

  • ADM2780Schem.png
    ADM2780Schem.png
    940.8 KB · Views: 33
cool!

yeah i did have it backwards, will do another inductance test on the primary this time,

i wonder if anybody has that discrete module reversed yet?

 
this is turning out to be quite a unique transformer,

big ol fat stack of M6, probably with low turns,

no wonder it took some excitation to get the core going again,


the can is aluminum

 

Attachments

  • adm a.JPG
    adm a.JPG
    112 KB · Views: 47
Wow, you got a lot of info on this thing!

Here is pri ind vs level,  as would be expected, saturation occurs at about half the voltage level as the 1:2 ratio means half the turns on the pri which means twice the flux,

you might try energizing your transformer with 20 Hz at 4 volts for a few minutes, then re-measure inductance, it would be interesting to see if you got the same results i did with the increase
 

Attachments

  • ADM OPT pri ind vs level.png
    ADM OPT pri ind vs level.png
    8.2 KB · Views: 13
potting compound is flexible type and comes apart with light heat,

lams might be 50/50 Ni alloy so you do not want to use stiff epoxy on nickel due to magneto-restrictive forces (lams expand and contract slightly with H field)


 

Attachments

  • a1.JPG
    a1.JPG
    174.4 KB · Views: 34
no wonder i could not heat up the bottom well enuff to lossen epoxy, it was not epoxy at all but G-10 circuit board type fiberglass,

weak point on this xfmr is that there is very little slack in the hookup wiresto PC board pins,
 

Attachments

  • a2.JPG
    a2.JPG
    131 KB · Views: 33
Back
Top