[ACMP investiupgradifications] All things PREAMP

[crazydoc]

I noticed the photo's from above have what appear to be different zoebel values. Any additional info there on the cap value? I don't have a cap decade box but could order parts to make one to do the tests. If you have the info on the zoebel it would be much appreciated.

There are two networks on the mic transformer.  There is 470pF + 12K1 in the position where the Factory put 1000pF and zero Ohms.

That damps the response quite nicely on the low gain ranges.  However, the high gain range has some fairly severe ringing with that same network.  I added a second damping network across the input to the first stage so that at high gain settings both networks are loading the transformer.  The damping network required to kill the ringing in the high gain mode was really severe.  It had considerably less bandwidth than the results on the low gain range.  I tried lots of combinations to try to get close to the same bandwidth as the low gain range, but was at first unsuccessful.  You could have either extended bandwidth with ringing or reduced bandwidth and no ringing.  Somehow I stumbled upon the idea of using a 2nd capacitor in parallel with the resistor and low and behold!  The bandwith came back without the ringing!!  In this case I was able to eat my cake and still have it.  The network across the input to the first stage is 2200pF in series with the parallel combination of 1000pF and 3K01.  The bandwidth is very similar to that of the low-gain range.

 

[0dbfs]

Thanks man! Are the line-in damping networks stock?

This is what I came up with for the original switch positions (in mic-mode):

[color=#0000ff]Switch-Pos      Mic-Trafo       Attenuation     Stage-1-Gain    Attenuation     Stage-2-Gain    Stage-3-Gain    Output-Trafo    TOTAL-GAIN-dB[/color]
[color=#0000ff]1               12             -31              0               0               28              11              4               24[/color]
[color=#0000ff]2               12             -25              0               0               28              11              4               30[/color]
[color=#0000ff]3               12             -18              0               0               28              11              4               37[/color]
[color=#0000ff]4               12             -11              0               0               28              11              4               44[/color]
[color=#0000ff]5               12             -3               0               0               28              11              4               52[/color]
[color=#0000ff]6               12              0               0               0               28              11              4               55[/color]
[color=#0000ff]7               12              0               25             -23              28              11              4               57[/color]
[color=#0000ff]8               12              0               25             -17              28              11              4               63[/color]
[color=#0000ff]9               12              0               25             -9               28              11              4               71[/color]
[color=#0000ff]10              12              0               25             -3               28              11              4               77[/color]
[color=#0000ff]11              12              0               25              0               28              11              4               80[/color]

Pretty interesting. I often use other pre's with gain settings as low as 6 - 12 dB depending on the source/mic.... In the originals 1R28 on this scheme (fixed@120R) is switched between 120/330 in the -50 and -45 sensitivity positions. It would be interesting to incorporate some of that switching capability into this circuit. The way it stands now, the 120R in the second stage is omitted in line mode, and switched to GND in mic-mode... I'm thinking a 2.5k pot or thereabouts mounted to the front panel would provide some additional tweaking of the second stage gain while keeping only that stage active for a particular switch range... Then maybe something similar for when second stage is active in place of 1R6 (91R).


Adding an OFF position to the 11 position sensitivity switch in mic-mode @ position 6 prevents the feedback-pop. Then there would be five sensitivity positions with only stage-2 active, and five sensitivity positions with both stage-1 and stage-2 active... Swapping 1R6 & 1R28 with a front panel pot would allow different gain settings for each of those five positions. Sort of a combo of the JLM 1272 hotrodmod and more standard 1073 gain stuff... Basically providing some control of how hard each gain-stage is driven and amt of that particular NFB making stage 1/2 variable between say 18dB and 40dB of gain. It would of course be possible to blow up the signal in certain cases or settings.


Prolly change the trim-pot between stage 2/3 to a 10k instead of 5k...


Since I am removing the LED overload indicators in these units I may be able to just-make-space for a couple more control pots...


Cheers,
jb

 

[mitsos]

My plan with the gain switch was to get one of those chinese 24P switches, wire it like the original and stick it in place of the one that's there (maybe dremel off the front of the PCB since it will all be wired in). I have no time to even think about it, but I think that would solve all the gain issues.

good luck with your units. Thanks for sharing the zobel info crazydoc!

 

[rattleyour]

Finally had a chance to use my 73s for the first time the other day-- albeit just on a single room mic.

I was surprised by the amount of insertion loss when the EQ is engaged, it's roughly similar to a -10db pad to my ear, but I'm just guessing.

I've read through this thread a few times over the years and, while I feel like there might have been reference to this, I can't find it.

Can anyone comment?

 

[crazydoc]

Are the line-in damping networks stock?

Here's what Steve wrote me while he was doing it:

"I have found after many iterations that the best results over the entire line-in range is to use a 1k resistor across the line input xfmr secondary.
It seems better than any RC network.  The resulting input Z is still about 10K so it seems a good solution."

However, in the pic there is definitely a cap there at the 1C21 position. I'll open it up again when I have the time and have another look.

 

[crazydoc]

Here's a pic of the line in network. It looks like 1K across the secondary (in place of 1R56), paralleled by a 470pf cap and 432 ohm resistor in series in place of 1C21.

 

[okgb]

Thanks Doc ,
Ah not sure why it didn't occur to me earlier, BUT
the ACM micpre thread inf compilation is now in groupmail DIY

Not 100% current , it should be dated, but thats the bulk of it
save for some recent stuff

 

[0dbfs]

Thanks Doc! This is great info to know.

Cheers,
jb

 

[chris carter]

I had a pair of 81s that were the first ones into the repair station.  After a very long wait, I found a kind soul with an extra pair of NIB 84s and was willing to trade for the rights to my 81s when they leave the repair station. 

Both 84s had some minor hum w/o EQ and some nasty buzz when certain EQs were in certain positions.  Did the toroid twist and then added 0.01" mumetal.  Made very clean circles with zero gaps using a pill bottle and a travel mug and electrical tape LOL.  4x thick for the toroids, 3x for the mid-band inductor, and if I remember correctly 2 for LPF & HPF (was getting tiny hum in those as well).  Coiling technique and thickiness provided a big variable in effectiveness.  The results were very good and more than usable for anything except delicate stuff + heavy compression with EQ using the lower couple mid bands with lots of gain.  Instead of running these with the output fully open, I'm going to run them with the output halfway down (I will at some point calibrate them and put a little white dot to mark the position) as the noise (and hum) get lowered and they have plenty of headroom on the front end.

And I must say, the do sound pretty good, cheap parts and all.  Pulled up some old sessions and swapped them out for 1084 plugs and was rather suprised.  Maybe not better, but definitely not worse either.  Mic'd a few things up they sound suprisingly good.  Only thing I might be careful of is soft vox that will get tons of compression with the EQ engaged because, not suprisingly, the noise floor does come up a bit, but that's not exactly a shocker. 

At some point I will remove the gain stage with the pop.  I can do without a gain position.  I have nothing more than cable making skills so I'll pay someone unless I acquire those skills between now and then.  I don't do a lot of recording (mostly mixing) so I'm not so concerned about them frying anytime soon as the gain knob doesn't go past that point very often.

 

[tommypiper]

Here's a pic of the line in network. It looks like 1K across the secondary (in place of 1R56), paralleled by a 470pf cap and 432 ohm resistor in series in place of 1C21.

Where is the pic?  Can't see it...

 

[crazydoc]

Here's a pic of the line in network. It looks like 1K across the secondary (in place of 1R56), paralleled by a 470pf cap and 432 ohm resistor in series in place of 1C21.

Where is the pic?  Can't see it...

It's working now. Anyway, on the schematic http://recordinghacks.com/pdf/tnc/1081-Amp.pdf , in place of 1C21 (2200pF) there is a 470 pF 630v WIMA in series with a 432 ohm resistor. In place of 1R56 (2.2K) there is a 1K resistor.

 

[Mylithra]

Finally conquered mine. When I first got them , I didnt have the knowledge or experience to fix them so I put them away. About 10 mic pres, 6 tube amps a pair of compressors and a pair of eqs (and a couple of years) later, I pulled them out and took another look and realized I could fix these now.
The biggest change I made was removing the power supply from the chassis altogether. I bought a small Hammond aluminum box, built 2 new power supplies. Upped the voltage rail to +/- 15v, then included the 48v and 24v taps. Pulled it down to 5 wires and ran the power via a 5 wire cable on a 4pin switchcraft XLR. (4 pins for the power and the shield for the ground)
There is now no hum, crackle, sizzle, or anything of that nature in the sound anymore.
(There is also no thanks to Steve H. whom I sent a PT to one of these units for him to tear up. I never got the replacement he was suppose to sell me when they were available But thats another thread altogether)

Got rid of the pop at the 2nd gain stage engagment by crossing the back leg of 8R15 with 8R16. This turned that particular switch position in to an off position but its better than what it was.
I also made small cups and covered them with a couple of layers of copper tape to make a shield for the inductors.
All these things have made these quiet enough to use now.

As for the harmonic distortion these things have, I consider it just another color on the pallet. If you think about, the reason we love Tube, old analog etc is because of the colors that kind of equipment imparts on sound. That color is harmonic distortion, but again, thats another thread altogether.

 

[clintrubber]

Just wondering if anything happened after Aug 2011 w.r.t. these pre-amps. Did a search for eventual new threads, but didn't found anything. All of these under a thick layer of dust ?!  8)

 

[mitsos]

where've you been man? long time...

for me, august 2011 was a bit rough, I needed cash and I sold my china84's to a local studio. I had done the gain mod, zobel reworks, replaced the lytics with tants for the sake of keeping it "original", and did various PSU mods scattered throughout the thread.  Nothing major but IIRC one unit was hum free, the other still had a bit of hum with the low-mid band engaged.  That hum was the major issue with mine, no idea why one was quieter than the other.  Maybe one passed QC?  ;D

 

[clintrubber]

Hi,

Apologies for that  , last years nearly all DIY-time was spent on ehh, m*s#c (band moved into writing soundtracks & playing these live with silent movies, pretty enjoyable)

And then this nice spot on the interweb comes to mind again, always nice to have a look.

Those preamps left pretty untouched/unmodded so far here, but I'll keep them.

There'll be a certain benefit in it though to have sold them, end of hassle  8)

Bye !

 

[okgb]

Nice to see you back !

One out of three I bought was use able ,  I did the xister mods , shielded the inductors
but I don't trust them to use , so they'll never be on the top of my use list and now there
are GOOD neve type kits and alternatives , they're heavy don't fit well into a rack and don't even
look that good .  Alex seemed to have made his work , I was going to replace the transformer ,
[ in fact the original idea was using it as a platform to upgrade the parts ] But Harry Vorderbrug
never sent me the xfmrs I paid for , so I'm not putting anymore money into mine if they're never
going to be something I use

 

[0dbfs]

Ha Ha! Mine all work well now. Used OEP cans for the inductors (chrion's 1081 inductor in one unit), swapped transistors, gain-mod.... They're nice.

I don't do much tracking or recording in my home studio but do at work where we have other pre's.

I use the acmp for summing box make-up-gain from time to time. Either that or an API 3124.

Take the tops of and they fit into 1RU but that's gotta be one of the worst production mistakes.

Ciao,
jb

 

[clintrubber]

Nice to see you back !

Thanks, good to see that 'most people' are still here 

One out of three I bought was use able ,  I did the xister mods , shielded the inductors
but I don't trust them to use , so they'll never be on the top of my use list and now there
are GOOD neve type kits and alternatives , they're heavy don't fit well into a rack and don't even
look that good .  Alex seemed to have made his work , I was going to replace the transformer ,
[ in fact the original idea was using it as a platform to upgrade the parts ] But Harry Vorderbrug
never sent me the xfmrs I paid for , so I'm not putting anymore money into mine if they're never
going to be something I use

Sorry to hear that, not a nice story and not an isolated story. 

So far my own needs the last years have more been w.r.t. portability, so I've been using other preamps (more channels in less rackspace).
But still living in the naiviety here that these ACMP-boxes will become brilliant stuff after a few hours with a soldering iron    We'll see.

Bye!

 

[alexc]

My pair of 81 are still going strong - they are certainly high on my list of most used preamps for mics.

No maintenance issues at all - they still sound and measure to a very reasonable spec installed in a large rack with lots of stuff all around switched on.

No doubt, if I were in the market again, I'd make a beeline to Colin's ez-1073 or Bluzzi's ez-1083

I think I paid around 425usd all up each including the transistor, inductor+shield replacements.
For the money they are unbeatable and the learning was definately worth while.

And what a SAGA for most  ;D  So many, many issues; technical, historical, evangelical and ultimately even a morality drama.


Makes me pine for the old days!

 

[alexc]

Just while on the subject of that long ago group buy, that tnc apex460 style mic ended up being quite a good mic, after reworking the circuit, tube and capsule. I think that came out at around 450usd.

I use that mic into one acmp81 feeding a diy la2a and it remains probably my best/most comprehensive mic chain.

Again, that led to a huge amount of learning as well as lots more time in focussed listening to the mic - indeed mics in general.

So, there is definately a 'happy ending' for at least some of the tnc stuff.