[ACMP investiupgradifications] All things PREAMP

[rodabod]

I'm guessing 50Hz as they use PAL video.

 

[alexc]

Hi Guys

I'm on 50Hz here in Oz. The most noise is at 50Hz and 100Hz decreasing at 150Hz and 200Hz.
Just starting up the PSU replacement.

Using a single (non-China!) potted toroid 50VA 24-0-24VAC  feeding (in parallel)  2x JLM Powerstations
each giving DC rails of +48V, +24V, +/- 12V and one negative rail unused.

A 5 core cable (rated at 2A each core) on 5-pin XLR socket out of an external rack from each
of the Powerstations to each of the ACMPs. I have to check out the grounding and figure out
the best way to do it.

Just received the parts today, including BC441s and BC461s (non "-6" variants, I'm afraid!)

Should have something tested this weekend, I hope. I'll post my findings as soon as I have something good happening.

Actually not so much of a hardship, as I already am building a 2RU 'DI' box to feed these ACMPs.
Uses JLM FET DIs (and also has a couple of Edcor based reamps and  Haufe based DOA style additional DIs for fun).
So, I just upped the VA of the toroid and add a couple of Powerstations and connectors.

With a bit of luck, these puppies will sing yet!

As a preamp, they sound pretty good. My normal reference mix (Stone Roses Second Coming)
sounded pretty full - no obvious frequency response problems. Knobs all worked fine.

Also, received the ACM1200 and other pre. The mic looks reasonably good - some
denting on the top of grill as others have noted and the pad/lo-cut switches pretty nasty too.
But the rest looks good. Will be burning it in and testing this weekend.

I tell you - I'm just happy to finally start moving to a more familiar DIY phase of this project.

Thanks to Chance - I believe he really worked hard to get this GB done.
I wish him luck in future GBs and thank him for his efforts.


See you

 

[okgb]

thanks for the nice detail, i haven't seen the schematic lately
but you figure a 5 conductor cable will do all the dc rails ?

D.I. ahead of the preamp input xfmr , nice option to have

 

[antichef]

I'm compelled to ask: ...

My compulsion got the better of me and I went ahead and rewired the one lead of the existing resistor at R43 to the leg of R44, and it seems to work for me -- thanks again!!

 

[0dbfs]

I'm compelled to ask: ...

My compulsion got the better of me and I went ahead and rewired the one lead of the existing resistor at R43 to the leg of R44, and it seems to work for me -- thanks again!!

Awesome!
I put a little heatshrink around the end of mine to prevent intermittent connections with the previous solder pad.

Cheers,
Jonathan

 

[okgb]

Hey Antichef [ what'd food ever do to you ? ]

it's be good if you referenced r42 thing as it comes out of context
are you talking about the gain switch ?

tia just trying to get ready for mine when they come

 

[antichef]

Hey Antichef [ what'd food ever do to you ? ]

it's be good if you referenced r42 thing as it comes out of context
are you talking about the gain switch ?

tia just trying to get ready for mine when they come

sorry - that was indeed a few steps removed from 0dbfs's original post:

http://www.groupdiy.com/index.php?topic=27791.msg381731#msg381731

yes - the gain switch.  I did it to both of my 81s and both of my 73s, and I'm thinking the numbering may have been different on the 73s, but the layout was the same. -- anyway, here's my picture, fwiw:

(I like food...)

 

[TomWaterman]

Because the value for customs was around USD 400 for the one pre. (much inflated)

Really 400USD for one preamp? Thats more than excessive, good work on the other fixes here though guys.

-Tom

 

[0dbfs]

Using a single (non-China!) potted toroid 50VA 24-0-24VAC  feeding (in parallel)  2x JLM Powerstations
each giving DC rails of +48V, +24V, +/- 12V and one negative rail unused.

A 5 core cable (rated at 2A each core) on 5-pin XLR socket out of an external rack from each
of the Powerstations to each of the ACMPs. I have to check out the grounding and figure out
the best way to do it.

Hey AlexC,

If you get a chance I would like to know whether you still experience hum with your potted 50VA 24-0-24 toroid inside the chassis. Not sure if that's a test you can do easily while you're at it or not.

If it turns out that a diff toroid in the case with a dedicated 5-rail-JLM-Power-Station works well, I might go that route instead of remoting the original toroid..

Thanks in advance!

Cheers,
jonathan

 

[rodabod]

Because the value for customs was around USD 400 for the one pre. (much inflated)

Really 400USD for one preamp? Thats more than excessive, good work on the other fixes here though guys.

It might be nice if Chance could sends ours as "going to be repaired" or similar in the tax declaration seeing as they don't bloody work! Or maybe there still is another way... We could always pay a US middleman to forward them!

 

[playboss]

as pennyless i am i already talked to him about this and he said to remind him once the green route is prepared  . I used to buy <50 usd rack gear from usa ebay and it comes thru without hassle, but the seller has to indicate it . So I guess its ok to tell him only once if yo have the greenie.

 

[alexc]

OK - so I have 1st mod results for the ACMP-81

- changed Q4,Q5 on 4 EQ bands to bc-441 and bc461 (non "-6" variant)
- replaced toroid with my usual China 30VA 24-0-24VAC unshielded
- replaced PSU with JLM Powerstation (+48V, 24V, +/-12V, Gnd)

My results are :

- basic noise floor of my converters < 90dBu 50Hz harmonics. Bal In/Out to AD/DA.

- with EQ OUT, noise floor slightly higher than before but 50Hz harmonics all < -90dBu 
  very little (almost none) added noise at all gain settings

- with EQ IN, all bands 0 gain, OFF frequency  : very little change to 50Hz harmonics   

- with EQ IN, Hi and Lo filters ON, all frequencies: very little change to 50Hz harmonics   
                  Hi      band ON, Max + gain, all frequencies  : very little change to 50Hz harmonics, <-90dBu mid hash   
                  Mid-Hi band ON  Max + gain, all frequencies  : very little change to 50Hz harmonics, <-90dBu mid hash   
                  Mid-Lo band ON  Max + gain, all frequencies  : 20dBu of 50Hz and mid-freq hash added 
                  Lo      band ON  Max + gain, all frequencies  : very little change to 50Hz harmonics, <-90dBu mid hash 

- with EQ In, all good filters and EQ bands in and max + gain, all frequencues : < -85dBu 50Hz harmonics, < -90dBu mid hash  ;D

So - nearly there. Seems internally mounted toroid + new PSU + replace transistors is *almost* great.

Now to work on the remaining offending Mid-Lo EQ and then try out externally mounted toroid + PSU.  :-\

See you

 

[alexc]

Should have added :

- haven't checked 2N3055 bias
- was done at unity gain for a 'normal' level of about -4dBFS on Cubase meters
- didn't change Grounding scheme : case grounded at XLRs and JLM PSU 100R+47uF between chassis and signal star

 

[rodabod]

Thanks Alex, that's really helpful.

 

[alexc]

Glad to help (after taking away so, so much from this forum    ;D)

Further futzing about on the offending Mid-Lo band shows that
the amp section is not noisy - definitely the filter part.

And it doesn't look like 'regrounding' the section does anything,
so I think it is induced noise.

 

[alexc]

Another part of the puzzle

- subbed out the offending lo-mid band board with one from my second acmp-81
  No change to the noise

- swapped position of mid-hi and mid-lo boards and the noise moved with the board
  There was still in the mid-lo position a mid-hash at -85dBu but all the 50Hz gone
  The Lo-Hi position now showed the 50Hz noise and higher mid-hash

So it looks like :

- the Mid-Lo board 50Hz harmonic noise is not position dependent but inherent to that board design
- the Mid-Lo position has some inherent mid-hash noise induced

Also, putting a 12ax7 tube shield on the inductors reduces mid-hash  by 3dBu (mid-hi) and 6dBu (mid-lo)

OK. Now what? Try to debug the Mid-lo board ? Check layout for bad or misplaced parts?
Look into replacing the inductor?

Now to externalise the toroid and PSU or 'learn to live with some noise in the mid-lo band'

That's it for tonight!

 

[rodabod]

Mid-Lo band ON  Max + gain, all frequencies   : 20dBu of 50Hz and mid-freq hash added   

By the way Alex, did you actually mean +20dBu? Ie, your converters were either close to or clipping? Or did you actually mean that the noise floor rose by a factor of 20dB?

 

[alexc]

I mean an additional 20dBu to take the noise from around -90dBu to say -70dBu.
Not close to clipping.

 

[rodabod]

I mean an additional 20dBu to take the noise from around -90dBu to say -70dBu.
Not close to clipping.

So, an additional 20dB with respect to the noise floor; I think +20dBu confused me as dBu is a unit of voltage level as opposed to a ratio. That's maybe not awful, especially since we might not take that boost frequency to maximum often. I do like things to be working as best as possible though!

 

[alexc]

One more bit

- subbed in a mid-hi board from second ACMP81 into the mid-lo position.
 So I now have 2 mid-hi boards installed in the mid-hi and mid-lo positions

 No 50Hz harmonic noise at all. And the overall mid-hash dropped a couple of dBu too.
 That rules out basic position dependent induced noise.

 It's either the layout differences between mid-lo and mid-hi that is the problem
 (they are about 90% the same excepting the filter section differences)
 or it is the inductor.

Tomorrow I'm going to :

- test the second ACMP without the toroid and PSU changes but with the transitor mod
- swap a mid-hi inductor into the mid-lo board to see if it is the inductor
 or the board layout of the mid-lo that is the problem
- do the external PSU and toroid

Now I'm going to bed.
I have to deal with real world instrumentation system QC issues tomorrow
(the ones that you get paid for!)