Another Amek Tac Scorpion overhaul!

[Eliani]

As this "vintage" console had been sitting in my stuffed workshop for a while, I thought this would be a good time to start a new project. I acquired the stinger in a trade for some awesome 500 series mic pré-amps I built. It smelled of cheap cigarettes and beer when it entered my house. And surely after a few days I regretted swapping my OV560's for this murky scorpion! But, when I fired it up it worked! I swiftly went through the console and found some scratchy pots and faders, a blown LED and some missing knobs, but  overall it looked in good shape. So I gave it a quick scrub and lifted some of the channels en for what I could tell it seemed not to have been worked on all that much, so I put it away for a time like this!

I've been researching a bit about a possible revival of this little console, and I found a great deal of information about it on the old analog console forum and some other forum, something with sluts? But I believe it might be a good dead to get it in proper working order before I start messing around with it. So I stripped her down to her bare bones, and made a list of things that her previous owner put her through, starting at channel one onto 32 and on to the auxiliaries, sub-groups,  and finally the master section. I had a small stroke after reviewing the list of stuff that had been done to this console, besides the relentless second hand smoke abuse!

But onward we go, and today we cleaned her out, gave the old girl a sponge bath, and really dug in th.... wait, this is starting to sound like a grannies porn movie!! Let's get to the facts.

We'll start with recapping the S4000 master section, well I find it hard to call it that as it's no bigger then a single channel strip, but the master section it is! The S4000 we've got seems to be worked on, and at that I believe it was done by someone who could read, but not interpret a plan as the coupling caps of 100uF 10v caps were replaced with 10uF caps right in front of the insert point! and the same happened right in front of the main outputs, but with 47uF caps. Now I could be stupid, but the only reason I could think of these would have been replaced by smaller then original parts could maybe be because they installed sowter 3612 transformers to balance the outputs. But I thought that the output impedance was 180 Ohms, which would be no problem for the 600 Ohms a Sowter 3612 carries on the input!? So to give you an idea of what was done to this master section, a list:

- C27, C28, C29, C30, C33, C34, C58 and C59 were replaced with wrong size caps, some of which I get.
- U11 and U12 , behind the fader were replaced with different brand TL072's .
- R75, R76 and R77 have ben cut from the circuit, effectively cutting auxiliaries 2,3 and 4 away from the talkback output!
- All ribbon connectors have been replaced with other connectors, so the don't fit right!!
- At least three traces or pads have been cut by de-soldering.

Anyway I think it would be best to revert this one to original to start with! I'll be replacing the caps with those little red ones Colin at Audio Maintenance sells. So he received an order for the following caps, together with 4 replacement pre-amp kits for the S1000 input channel., and within a few hours I got the message the were already on their way!

As I progress through reviving my little scorpion I wouldn't mind everyone's input, I'm sure a few of you have done this before and maybe some of you will do so after reading the end result of this trip.

A quick measurement taken through the masters insert return channels of the left and right mix-buss. And the result: +/- 0.5dB from 20Hz to 20kHz. And the fase was inverted because of the inverting fader booster amp.

 

[joaquins]

Nice one!!!

I think you should try measuring at lower frequency to see if there is a peak, that low end bump may end in a peak under the audible range caused by a resonance between cap and transformer or something like that which may bring some problems...

JS

 

[Eliani]

That low end bump may end in a peak under the audible range caused by a resonance between cap and transformer

My measuring equipment doesn't allow me to go much deeper then 10Hz, and then it's not very stable around that region! I use only my system analyzing software (Smaart v7) for these measurements. Is there a low-cost alternative for this kind of measurement? something that can do THD too? I don't know really!

 

[jackies]

Well, the old http://audio.rightmark.org/products/rmaa.shtml is rather low cost.

 

[Eliani]

Jackies pointed me to this nice little app that I remember using before, but never to much use. But I ended up downloading the app, and run it as a test on my M-Audio fast track r8, I've rented out the Andiamo, so the m-audio will have to do!

Now I know that if your measuring THD, with or without N, through your computer, or any equipment for that matter you just need that equipment to have a better THD+n than the equipment you want to test. So first up, testing the m-audio in/out. This test gave me a THD around 0.0017% slightly better than what they stated in their manual!

But the fader and fader amplifier of the scorpion alone brought that down to 0,0082% , the rest of the measurement doesn't make much sence to me yet!

 

[Eliani]

So I've been reading up on my little console, and I got a bit wiser, more confused, but ever so determined to get this show on the road! I've ordered a master section recap kit (although I ordered some radial caps that should have been axial caps...) and 4 langley mic preamp-kits from colin at audiomaintenance.com. I've bent my mind over the options and came up with a plan! we'll see, I deviate easely...

So stuff I would like to do.
* Rework that master section, revive it as you would say!
  - Recap with close to original value's
  - Replace R56 and R57 with 22.1k resitor for the "unity gain master fader mod"
  - Try LME49860NA, TLE49720NA, BurrBrown OPA2134 chip in various places.
  - Remove C23 and C24 (replace with ferrite bead)
  - Remove or rather short/bridge C27, C28, C30, C33, C34 and see if this improves the overall sound
  - Add 0.1uF mono ceramic caps from pin 4 and pin 8 (+/- 17V) to GND on all opamps  to improve common mode rejection.
* Recap and swap the mic pre-amps with Langley design pre-amps for 4 of the 32 channels!
* Recap and swap the mic pre-amps with Langley Design line-amps for 4 of the 32 channels!
* Totaly overhaul/pimp/spoof and add bling to 2 or more channels for vocal tracks.
  - Add fancy input transfomers for debalancing.
  - Rewire direct-outs and insert points, possibly selectable.
  - Create 2520 DOA based pre-amp (something like a VP26?)
  - Or skip this and add more line-level amps and use my CAPI pre-amps instead!
* Recap the remainder of s1000, s2000 and s3000 channels.
* Have a beer!
* Rethink the metering because there is nothing original about that thing!
* Fit the desk in a proper desk, something fancy, something elegant...

 

[gar381]

Hi Eliani

I recaped my Angela some years back so I know you are having FUN!!
I did!!

BTW.. I played around with opamp swaps and kept all the original guys.
Just liked the way the originals opamps ( TLO7x and 553X) sounded with
the AMEK circuits.

I'll be watching this thread with much interest.

GARY

 

[mitsos]

Do the PSU first, if that's not done already. Then test again.  The Scorp PSUs sometimes develop intermittent problems because the regulators are socketed. Audiomaintenance has a nice PDF on how to mod the PCB to remove the sockets.

I'll be doing mine soon (I've said that a few times already).  Definitely have fun swapping opamps and things, U5 in the input channels could be changed to something with more beef,  though for the master I'd recommened using an ACA-BO from classicapi to replace summing and booster opamps (U11 and U12 IIRC).  LME and others did not sound the same even with output trafos.

I've  tried several line circuits so far vs the stock, and came up with one I like best, I posted it in a thread somewhere here, and installed it on the stock PCB with jumpers and strategic placing of components. But I'd still like to keep mic/line capabilities on modded channels, so I started designing an "API" type input mod PCB but it's half done on my computer somewhere..  There was another member critter something that was also doing some mods to his, not sure what happened to his project (mine's slowly coming off the back burner, eventually!). 

I'd be interested in the metering changes. What do you have in mind?

 

[Eliani]

Though for the master I'd recommend using an ACA-BO from classicapi to replace summing and booster opamps

So you would, or already have replaced your mix-bus and fader booster circuit with the 2-ACA-BO by ClassicAPI?
That's an interesting approach, but a more expensive one with 4 DAO's, 4 transformers and then some! I'll keep this one in mind next time I order from Jeff Steiger! Damned Mitsos, you've got me deviated already!! and yes, U11 for left mix buss and booster, U12 for right mix buss and booster!

Looking at my original manual, which was in great shape when I got it together with the console, on the last page I find the folding pages containing the schematics of this console. While looking at the bus mix circuit, there's nothing their that warms my heart... just the look of that single TL072 channel being used  as an inverting opamp (which explains why my insert return came back inverted!!)

But, first things first, We'll have a look at that PSU, and revert this master section to it's original state!

 

[mitsos]

Yes, I have used this and it works 100%.  Others have used it in other consoles here too, if you look around you'll find posts.  Basically it can go into most any console that uses the inverting opamp summing amp + a booster.  Most stuff from this era is probably like this.  The ACA does the same exact thing as the TAC circuit, just much better.  If I were you, (I'm not so feel free to ignore of course), I wouldn't spend too much money on the master as it is, and just save up for the ACA.  It's that much better. Just recap it and stick some LME49860's for now and leave it alone.

There are some small errors on the schematic that had me going in circles trying to install the ACA.  IIRC all the left channel R's and C's should be odd numbers and right channel should be even, like we do with L+R pairs. But on the schematic,, some of the even/odd R's and maybe C's are switched around. Doesn't make a difference if you leave it stock because the PCB is correct obviously, but if you try to mod the circuit, it could drive you crazy trying to figure out why your left channel is coming up on the right monitor.

But, yes, stop everything for now and do the PSU first.

 

[Eliani]

I'm trying to decide which channels I would upgrade with the Langley preamps? I can choose between 26 out of 32 channels that look original to me, nothing recapped, or messed around with, or should I choose to revive one of the channels that has been messed with? One could say to do the former, and trusting that the guy before you did a decent job. One could not trust this guy, which I don't, and take one of his messes and straighten things out?

I consider 4 or more caps the maximum for a channel to be piled upon the "not original pile" there are some channels that have one or two caps changed, but I can bring myself to believe that's normal repair! But I don't get this guy! While everybody seems to wanna put bigger caps in certain places to get better low-end results this guy seems to like to put smaller caps in wherever he can!

example:
S1000 module done by previous owner:
  * He changes the input caps (C2 & C3) with 22uF/100v where you would expect 47uF/63v
  * The Gain cap was switched, but by the same value.
  * C10 before the eq became 47uF where their should be a 100uF/10v.
  * C24 after the eq he left same value as original.
  * C25 coming from the insert return went from 100uF/10v to 22uF/63V
  * and it goes on like this..

Maybe this guy tried to open up the high frequencies a bit?

 

[audiomixer]

Maybe this guy tried to open up the high frequencies a bit?

why should altering the very low frequency response alter the top end? while the values used might still deliver reasonable LF response, you might want to check if there is a measurable influence at all... the coupling caps values will define a -3dB point together with the corresponding circuit impedance. seems there is a factor two at some locations, and a factor of four around the insert return. if you had a -3dB point around a few Hz, let's say 4Hz, you will get 8 to 16Hz -3dB points. that might be difficult to measure anyway....

- michael

 

[Eliani]

Maybe this guy tried to open up the high frequencies a bit?

why should altering the very low frequency response alter the top end? while the values used might still deliver reasonable LF response, you might want to check if there is a measurable influence at all... the coupling caps values will define a -3dB point together with the corresponding circuit impedance. seems there is a factor two at some locations, and a factor of four around the insert return. if you had a -3dB point around a few Hz, let's say 4Hz, you will get 8 to 16Hz -3dB points. that might be difficult to measure anyway....

- michael

It would not alter the high frequencies, you are right, but your ears would perceive it so.  And you're right again that even with 2 or more caps in series the cutoff frequency would be really  low, but truth be told, If you want a clean frequency response you try to get them as low as you can, it wouldn't only lower the -3dB point, it would lay the -1dB and -2dB further down the frequency curve too! And since we try to work with precision tools I would like it to be as precise as possible.

I revived a  couple of 200B consoles with the help of Jim Williams' writings and I love everything about it, it sounds more open, more punchy, more of everything, or so my ears make me believe it to be!! ;-) It's al relative!!

 

[audiomixer]

the stacked high pass filters all around in a console does indeed add up to quite a total filter, so the individual filter counts, sure. I just wanted to point out the metrics. I love the Idea of sub 5 Hz flat response for the total signal path, also because of the phase turn of a higher order filter. but I dare ask whether it is relevant.
If you look at the typical playout chain with in some cases 'protective' filtering of the signal in the tens of Hz region - and in many cases uncoordinated replicas... high pass in the controller, an other one in the amp...
I would love to see real world measurements of both cases. but getting precision measurements below 20 Hz can be challenging.
I did hear sublow extended speakers in the past (equalized for f-3 around 16Hz) and it is true that it alters the mid / high frequency perception. but your predecessors mods went in the other direction...

- Michael

 

[Eliani]

I would love to see real world measurements of both cases. but getting precision measurements below 20 Hz can be challenging.
I did hear sublow extended speakers in the past (equalized for f-3 around 16Hz) and it is true that it alters the mid / high frequency perception. but your predecessors mods went in the other direction...

- Michael

I know how hard it is to measure low frequencies, and I'm not even speaking about 20Hz! Everything below 120Hz is hard to get in line! I've been a full time live system engineer for over 15 years now, and still I face the difficult task of providing an even spread low-frequency image for an audience. But in my studio, and you should take that thing with a grain of salt, I believe in getting it right!

But my predecessor is responsible for the nicotine clogged pots, the half-assed metering bridge, the blue fader-plate, the bad soldering, and whatever comes next too... but I'll set it straight! even if it means to remove all electrolytic caps in the audio path!

And when I'm done I'll stuff it full of DOA's, transformers and what not, and ruin my work once more!!  8)

 

[audiomixer]

-off topic-

there have been quite a few new ideas over the past ten years or so regarding low frequency distribution in live sound, although dreadful L/R sub placement is still very common. the summing right along the center works great for 50 people and the mix engineer, but the rest of the venue get's frequency dependant LF crap.
stuff we like include sub arrays, directional subs, vertical stacks. but most of the time you get a classic stack under the PA.

- Michael

 

[Eliani]

-off topic-

there have been quite a few new ideas over the past ten years or so regarding low frequency distribution in live sound, although dreadful L/R sub placement is still very common. the summing right along the center works great for 50 people and the mix engineer, but the rest of the venue get's frequency dependant LF crap.
stuff we like include sub arrays, directional subs, vertical stacks. but most of the time you get a classic stack under the PA.

- Michael

Hehe, i get that too!!

End-fire setups, and logarithmic curved sub arrays are better, and in combination with flown line-source sub-arrays will outperform any setup!! But from time to time there is no space in front or under a stage to get an array in, so left-right it is. But even in a left-right setup you can do better than the old bulge... 

 

[mitsos]

the coupling caps values will define a -3dB point together with the corresponding circuit impedance.

Like Michael said, you'd have to figure out the minimum acceptable -3dB point at each location. Maybe the last guy calculated tha values he used, or maybe that's all he could find, or maybe he just didn't know any better.  I doubt there is a huge difference, even at the insert return, but yeah, I would go to the original values when you recap.  Important, use GOOD quality caps in the signal path, Elna Silmics are great, so are panasonic bipolars.  Nichicon has a whole line of audio caps, some of which (KZ, top of the line) I personally did not like compared to the Elnas for example, but that was a single, "controlled" test, so YMMV.

 

[Eliani]

Important, use GOOD quality caps in the signal path, Elna Silmics are great, so are panasonic bipolars.  Nichicon has a whole line of audio caps, some of which (KZ, top of the line) I personally did not like compared to the Elnas for example, but that was a single, "controlled" test, so YMMV.

I use Panasonic bipolars in all my "Iron Projects" and Nichicon PW (low impedance) for most console mods, but for this one I turned to Colin, I'm sure his choice will be best.

 

[mitsos]

I'm sure Colin stocks decent caps, but, personally, I would not buy hundreds of caps unless I had already tried them them in a couple of channels.  There are audible differences between different caps, some more obvious than others.  I only mention this because it seems like you plan on using this mixer in the future, and you could be doing yourself a disservice by not auditioning a few different types.