Another Amek Tac Scorpion overhaul!

[mitsos]

Before the kids go to sleep I managed to finish the API 312 input too! What to do with free time on your hands, and parts in transit!! So the AE2622 trannie, might replace that one with Lundahl/Cinemag footprints. Now all I need to do incorporate the line-level input, but I've got a feeling I'll manage that even with the original switch in place!

That's cool, keeping the original mic/line switch was my goal as well, before I got "sidetracked" by other projects.  I have to go back to the schematic and see how I did the switching, I remember I was scratching my head for a while.

 

[mitsos]

----- image removed ---- incorrect

Noticed you said your previous picture was incorrect. This is from Eagle, is it 0.1" spacing?  I want to get back to my PCB one of these days, so if you have the correct spacing it would help! thanks!


BTW, for the input trafos, you can get the EA/Cinemag/etc footprint, the Lundahl 1538 and the OEP to fit in the same space, I did it on the 312/M1 pcbs I did a while back, I think I posted an image of the traces somewhere. Just have to play with the rotation and they all fit, in case you want to have different options.

Also, you don't need the diodes, the S1000 has them at the main power input to the PCB, if you add them there you'll get another 0.7V hit.

 

[Eliani]

Noticed you said your previous picture was incorrect. This is from Eagle, is it 0.1" spacing?  I want to get back to my PCB one of these days, so if you have the correct spacing it would help! thanks!

The grid spacing on that picture was 0.05 inch, but it would be good if you would remove that picture from this thread as I swapped the V- and V+ pins on that board, and missed the connection on the caps too. I will post a corrected design as soon as I confirm the layout with the PCB's I ordered.

BTW, for the input trafos, you can get the EA/Cinemag/etc footprint, the Lundahl 1538 and the OEP to fit in the same space, I did it on the 312/M1 pcbs I did a while back, I think I posted an image of the traces somewhere. Just have to play with the rotation and they all fit, in case you want to have different options.

I did think about that, but the only transformers that have a low enough profile to fit between the channel-strips might be the Lundahl LL1528 and LL1538. To incorporate different input transformers it would be better to make the Sub-PCB bigger than the S1000 board and fit them on the bottom layer (mirrored). I stand behind this remark because I do the same for my "Iron Project" direct injection designs (www.eli-audio.com)

Also, you don't need the diodes, the S1000 has them at the main power input to the PCB, if you add them there you'll get another 0.7V hit.

For now I'm going to leave and use them as I do not want to burn other channels by miss-fitted components, I guess the 0.7v hit will not affect proper working of the circuit since the DOA's should work with only 16v.

 

[mitsos]

Noticed you said your previous picture was incorrect. This is from Eagle, is it 0.1" spacing?  I want to get back to my PCB one of these days, so if you have the correct spacing it would help! thanks!

The grid spacing on that picture was 0.05 inch, but it would be good if you would remove that picture from this thread as I swapped the V- and V+ pins on that board, and missed the connection on the caps too. I will post a corrected design as soon as I confirm the layout with the PCB's I ordered.

You mean the V+ and V- coming from the S1000 board right? The opamp looked correct.

BTW, for the input trafos, you can get the EA/Cinemag/etc footprint, the Lundahl 1538 and the OEP to fit in the same space, I did it on the 312/M1 pcbs I did a while back, I think I posted an image of the traces somewhere. Just have to play with the rotation and they all fit, in case you want to have different options.

I did think about that, but the only transformers that have a low enough profile to fit between the channel-strips might be the Lundahl LL1528 and LL1538. To incorporate different input transformers it would be better to make the Sub-PCB bigger than the S1000 board and fit them on the bottom layer (mirrored). I stand behind this remark because I do the same for my "Iron Project" direct injection designs (www.eli-audio.com)

I did make my PCB deeper.  It won't make a difference if it is mirrored, the distance between PCBs will always be the same, but getting them out of the frame might involve removing 2 at a time.

Also, you don't need the diodes, the S1000 has them at the main power input to the PCB, if you add them there you'll get another 0.7V hit.

For now I'm going to leave and use them as I do not want to burn other channels by miss-fitted components, I guess the 0.7v hit will not affect proper working of the circuit since the DOA's should work with only 16v.

It's your design, of course, I was just pointing it out because each S1000 channel already has those diodes on it at the power entry to protect it from reversed voltage rails. 

Anyway, good luck, please post back when you know the final layout of the pins!

 

[Eliani]

So we received the PCB's Rev.001 (no trimmer for CMRR, will be implemented from rev.003) as I figured I missed the pin-setting by 0.025 inch for the voltage rails. But no problem, bending those pins works for now. I've installed one and it works!! That's all I know right now.  We'll be changing the name for these PCB's to "El-252" as it has nothing to do with the 312 schematics..!

 

[mitsos]

Nice work. Now to figure out where to hang an output transformer!  8)

I'm actually feeling motivated to work on mine! Started prepping, counting modules to make sure none went missing (lost one input when I had to move, but got lucky and found one cheap on ebay). Hey I also pulled out the bags of knobs and counted them today.  A ton of the knobs are really bad looking, mainly the caps are scratched, faded, can't see the line any more, etc. I got a bunch of NOS knobs with grey caps, so between those and the ones I can salvage from mine *I think* I will have enough knobs, but many with terrible looking caps, which will actually be easier to ship, that is, if you're still willing to part with them.  Thanks in advance! 

 

[Eliani]

Nice work. Now to figure out where to hang an output transformer!  8)

I'm actually feeling motivated to work on mine! Started prepping, counting modules to make sure none went missing (lost one input when I had to move, but got lucky and found one cheap on ebay). Hey I also pulled out the bags of knobs and counted them today.  A ton of the knobs are really bad looking, mainly the caps are scratched, faded, can't see the line any more, etc. I got a bunch of NOS knobs with grey caps, so between those and the ones I can salvage from mine *I think* I will have enough knobs, but many with terrible looking caps, which will actually be easier to ship, that is, if you're still willing to part with them.  Thanks in advance!

My knobs don't look much better I'm afraid. But let me know what you need see what I can part with.

To get back on transformer placement, only way to fit one and still be able to remove a single channel would be to raise the sub pcb as high as possible and mount all components on the bottom layer. Although I do not believe it is worth the trouble to put input transformers in this console, at least it's not worth it to me. I think it be better to have good line-receivers and external 500 series pre's with transformers. This way I can choose different preamps fir different purposes.

But then that's my idea...

I'm going to order a new batch with the correct pin-setting and see if I can improve the eq-section. I received Jeff's 2-ACA-Bo last week and was thinking to implement the summing to a stereo pair of bus-modules rather then the master section, or build a monitor/external summing controller with it. Again for the sake of versatility...

 

[critterkllr]

You're fast!

My PSU will be coming in about a week, so I'll be back on mine soon. Although I still need an interface with more outputs.

I'm not sure if the Tac Scorpion I modules are the same layout as the Scorpion II modules. If they will work out with a little modification, I'd be happy to go in on some pcbs.

 

[Eliani]

You're fast!

My PSU will be coming in about a week, so I'll be back on mine soon. Although I still need an interface with more outputs.

I'm not sure if the Tac Scorpion I modules are the same layout as the Scorpion II modules. If they will work out with a little modification, I'd be happy to go in on some pcbs.

From what I can tell the PCB layout might be slightly different concerning the gain stages because the Langley mic-pre add-ons look different on the top for  Scorpion 1 and Scorpion 2 channelstrips. But looking at Colin's line receiver sub-pcb's they are both the same for S1000/S1100  and  S1001/S1101 channelstrips...

 

[Groovi]

Nice project.

Im very interested to see how this goes. I have two TAC Scorpions, one with 40+2/8/2 channels and custom meterbridge (or meterbox, i could say).

 

[Eliani]

IIRC this is how it's done.  Won't be able to test again for a couple of months, though.
Remove C4, C28, R55, and C34  (NOTE: R55 is probably supposed to be R54, this is what I meant that the schemo has errors, all right channel components should be even numbered, IIRC).
Then, using shielded cable with shield tied to ground on one side:
Circle 1 is where you take the signal to the ACA, Circle 2 is where it comes back to the master module so it can go to the insert and fader.  Circle 3 is where it goes to the booster and Circle 4 is where it comes back again. 

The signal goes to the TAC's monitor outs after the 47R output resistor, whether or not you use the ACA-Bo, so you'll have one more IC in your monitor path with the stock version.  If you want to monitor directly (you'll lose separate level control, but some people use external CRM systems, if so, then I think it's better to skip the ICs), you can mult the ACA-bo output direct to the monitor outs, (the ACA-Bo has two connectors per output), or use the other trafo isolated output, but your monitor would be at -6dB with respect to the main out).

Since we're waiting on the Rev_003 PCB's for the 2520 based line-receivers I built the 2-ACA-Bo I ordered from Jeff. I was just figuring out where to implement the board and went back to reading your guidelines. So this is what I got.

• Remove C23/C24 (point 1), this is where I could take the signal to the ACA-Bo. This point would be right after all channels bus resistors with a value of 10k each. I get this step, rather obvious.
• Remove C27/C28 (point 2), and return the signal coming from the ACA-Bo. This step seems a bit strange because the signal returning from the ACA-Bo is a differential line from the 2623-4 transformer, or did you skip this trani?
• Remove R54/R55 (point 3), where we pick up the signal coming from the fader and send it back to the booster.
• Remove C33/C34 (point 4), this is where we can inject the signal coming from the booster back into the original circuit.

Ok, so far I'm with you on this project. I'm probably going to wire as much as possible off-board, and leave the original circuit intact but remove the ic's for the original mix-bus and fader-boosters, always great for an easy revert. I'm going to use the EA2623-4 insert-pick-off as intended by Jeff. Balanced out, unbalanced back into the booster. I'll have to add, or steal a Jack-plug from another in/output but it's for a good cause!!

I'll need to figure out the best pick-off point to draw current for the opamps from.

 

[mitsos]

• Remove C27/C28 (point 2), and return the signal coming from the ACA-Bo. This step seems a bit strange because the signal returning from the ACA-Bo is a differential line from the 2623-4 transformer, or did you skip this trani?

Ok, so far I'm with you on this project. I'm probably going to wire as much as possible off-board, and leave the original circuit intact but remove the ic's for the original mix-bus and fader-boosters, always great for an easy revert. I'm going to use the EA2623-4 insert-pick-off as intended by Jeff. Balanced out, unbalanced back into the booster. I'll have to add, or steal a Jack-plug from another in/output but it's for a good cause!!

I'll need to figure out the best pick-off point to draw current for the opamps from.

Looking at it again, I would probably remove the resistor after C27/C28 instead, otherwise it will be in circuit all the time.

I wouldn't waste time with the balanced out, unbalanced in insert.  If you're going to do that, might as well stick a couple of your line receiver PCBs in front of the fader and have it completely balanced.  But you won't need it unless you have noise.  I saw Jeff's fader hookup diagram showing the trafo secondary going to ground and didn't think much more about it.  It was more important to install this so it works seamlessly as part of the mixer. 

any more progress on yours?

 

[Eliani]

Here's some progress...!!

I've actually installed the ACA in the console today! Only used the first pick-off point and wired everything off-board only to return  it after the booster amplifier. There's a lot I figured out as I went forward. Like replacing Rf and Cf (the later still needs to be calculated to achieve the -3dB drop-off point around 120Hz), I didn't even see that on the schematics diagram before I ordered all the parts, just like the resistors on the boosters input, so I ended up placing  two 100k's in parallel to get close to 51k as I didn't have  that value stocked. I cheated on Rf to, should have been 5.9k, but 6.81k was the closest I found laying around. So instead of the 0.59 gain I ended up with 0.68... still don't get if that's dB's and if it's a negative number?! And the fader response isn't what you could call logarithmic!

 

[mitsos]

The 0.59 is voltage gain. For every 1 V going in, 0.59 comes out.  In dB it is negative. 

Here is a cool online calculator:
http://www.muzique.com/schem/gain.htm

I don't know what you mean by the fader response not being logarithmic, is it not working properly? 

 

[Eliani]

I don't know what you mean by the fader response not being logarithmic, is it not working properly?

Between - 20dB and the top of the fader there's not much volume change...

 

[mitsos]

I don't know what you mean by the fader response not being logarithmic, is it not working properly?

Between - 20dB and the top of the fader there's not much volume change...

Look carefully at your wiring and compare it to the schematics and diagrams that came with the ACA-BO.  Look again at the booster one, there are a lot of options at the input, I forgot a link or the 51K resistor or something the first time.  You can also test that the individual amps are working properly by sending a signal to each, then measure the input and output and see if the voltage gain is what you calculated.  If you have a module extender you may be able to use the console oscillator for this, because you need to measure the input prior to the 10K resistor. You can use an ipod too, just need to stick a 10K resistor in series.

can you post a diagram of how you wired it?  You may even find your mistake as you draw it.

good luck.

 

[Eliani]

No report on the reversed logaritmic (or watherver curve) on the master fader, but...

Rev003 of the El-252 Differential input PCB is in!!

The pin-setting is still slightly off... but there's only a little bit of force to apply to get the PCB seated all the way to the main-board!! So now with DOA-sockets and CMRR trimmer...!! As Mitsos told us, the rectifiers on the main PCB are there for the same reason I put them on my sub-pcb, so no need to put them on, but I replaced them with 10R resistor in the same way Jeff does with his ACA's.

EDIT: If anyone needs some of these boards, I've got plenty!! Send me PM I'm sure we can work something out?

 

[Eliani]

I was able to squeeze three more in today!! But, I feel that the 25k trim resistors are not accurate enough to pin-point the right spot for excellent CMRR! It's way more easy to measure 1% resistors then the multi-turn!

 

[Eliani]

While setting up the for the making of 24 more line-receivers I was reading Jeff's CMRR documentation. While adjusting Rf for maximum clutter rejection I found it hard to set the 14 turns 25k to an exact reading. Would it not be possible to parallel a say something like 47k resistor with Rf this would give me a smaller range for setting Rf to cancel out common modes...?

EDIT1: It works as I thought, it's way more accurate when  I place a parallel resistor next to the trimmer. It's a lot easier to dial in the CMRR this way. Now for even greater accuracy I need to figure out a setup that gives me even better results, and I'm thinking there should be a perfect combination between two resistors, one trim-able the other fixed which gives me the full multi-turn value just to dial in the sweet-spot... Anyone?

EDIT1bis: I guess this was an easy one... it's better to get a 9.1k resistor and a multi-turn trimmer around 2k and in series and it's an easy fix!!

EDIT 2: So what about Cf? I read so much about this cap that needed to be changed in the ACA-Bo to adjust for the High-Pass filtering. Because in the original schematics for the Differential Amp Rf is 100k, should I need to change this one too?

 

[mitsos]

You're a better man than me.  I would just match 4 resistors and call it a day.

Cf is what? The feedback cap in the summing amp?  You changed Rf because your summing resistors are different from APIs, so now you change Cf by the same ratio in the other direction to keep the same upper frequency limit. (low pass, not high pass). If you used 6K8 like you said, you'd need about 138pF.  150pF is probably best, calculates to about 156KHz, other closest standard value is 120, that goes up to 195KHz.  Either one should be fine really.