TAC Scorpion Overhaul Odyssey

[Potato Cakes]

Less opamps can be cleaner,  how many series tl071 (072) can you tolerate before it matters? But if you go to a "better" opamp the problem diminishes.

The main issue with the s800 is that when the mid bands share an opamp their response becomes interactive, adjust the low mids and you affect the high mids, and vice versa. Is this cool or annoying?

I would say keep the current topology.  Instead focus on the other details.  Maybe different caps,  maybe different opamps, adjust resistors to get the Q you want, lower resistance when possible to reduce noise, add psu decoupling, etc.

I don't necessarily mind that the two bands would be interacting, I am mostly concerned how it will sound when boosting frequencies versus stock. This is where I will have to make the different alterations to a couple of channels and see how they actually sound. For now I a have a couple of ways to skin this cat. I think now I need to start making a component count and start getting a mouser cart started for at least the capacitors.

Thanks!

Paul

 

[Potato Cakes]

And another day off is spent in a hotel to stare at schematics and theorize about the possibilities of this project....

A friendly fellow in this community provided me with the schematics for the Trident TSM input channel, and it is obvious to even someone as electronically ignorant as I that this and the Scorpion are very close to being the same. Besides having different values, the major differences I see are there is an extra resistor from the output of the op amp associated with the frequency sweep to the "+" input of the buffer (or gain, I'm not sure exactly what it's technical purpose is) op amp, there is a capacitor parallel with the resistor connected to 0V to the buffer/gain "+" input, and there are two resistors parallel to the "+" input of the frequency sweep op amp which are also connected to 0V. Also, the TSM schematic has electrolytic capacitors separating the connections between bands, which are not present in the Scorpion. However the entire schematic for the TSM channel I have as all of the electrolytics circles as if they were slated to be changed or removed. I have noticed on a number of mods for various designs call for removing electrolytics in the audio path, but I am not sure why. And would they be replaced with a jumper or a film/ceramic capacitor?

It now becomes more apparent what has been pointed out to me by others that there is really nothing wrong with the design and that only some minor tweaks would be needed to "improve" the sound of this EQ. This leads me to more questions:

Would adding the additional capacitor parallel to the resistor connected from the "+" gain/buffer op amp to ground provide any electrical benefit (C30/R39)? Same for the additional resistor parallel to the "+" of the frequency sweep op amp (R45/R46). It is curious to note that on the TSM BOM R45 is 6k8 and R46 is 150k and I guessed these would be the same value.

For the cut/boost potentiometer, if I keep the stock 22k linear and use that with the TSM schematic values, will that shift the null/flat point in relation to the position of the potentiometer? Does the value of R58 determine this? It is 22k on the TSM and 10k on the Scorpion.

The BOM for the TSM used polystyrene capacitors parallel to the feedback resistors and they are ceramic on the Scorpion. Is this just because that is what was available when the TSM was manufactured (approximately 10 years before the Scorpion) or is there an audio benefit to this?

Will adding the extra resistor (R48 - 470R) between the frequency sweep and gain/buffer op amps in the Scorpion circuit provide any benefit?

I'm trying to gain more knowledge about what is happening on the component level and why things interact the way they do to aid in making better decisions before I actually start tearing into this console. The unwanted side affects are more questions and more headaches. Sometimes I wish I would have just remained ignorant and believed what the marketing guys told me as gospel.

Again, I am very grateful to everyone here lending me their brain cells to help with this massive project.

Thanks!

Paul

 

[Potato Cakes]

Just started learning AutoDesk and I did a quick schematic of the 553 eq as it would apply to the bus if this was done as an add-on card. This is my first go at this and I am struggling to properly label the components so they aren't overlapping with the symbols, but I just used the values straight from the 553 schematic that is readily available. It seems this could be a pretty simple board that would be easy to implement. Once the properly sized potentiometers and spacing thereof was determined to match the existing Scorpion metal work, this card could be installed inverted to the Scorpion card (with all old EQ components removed) and the new potentiometers would mount in the holes where the old ones went. It would just need connections for In, Out, V+, V-, and 0V. The original EQ IN switch on bus channel would still switch the circuit in and out. The other consideration would be to see how the DOA would fit. The easiest way would be to make the add-on PCB deeper than the Scorpion board to allow the DOAs to be installed/changed without space restrictions. The actual available space in the Scorpion frame would need to be measured to see if this could even be an option. Also, some resistors would need to be added to reduce the voltage from 17.5VDC to 15-16V so as to not burn up the the standard issue DOA. Or they would just have to be built with different components to handle the slightly higher voltage.

The other consideration is that I will have to adjust the feedback resistor to give me the desired output gain as I am not including the output transformer in this circuit. I think the extra weight of a 2503 type transformer will complicate mounting the board and put too much stress on the potentiometers that are holding the add-on board to the panel.

Once I get home I'll do some measuring and then I'll probably seek some help turning out some boards.

Thanks!

Paul

 

[jensenmann]

As you seem to have found out there´s nothing wrong with the Scorpion channel EQ. Since it´s easily tweakable let us know what you want to improve, then we can help.

Let me tell you a story about my former monitor desk, a 24 channel TAC 16/8 FB. Since my livesound days are over I had this desk sitting in the corner with 0€ resale value. So I decided to rack the channels since I really liked the EQ. The circuit is basically the same as the Scorpion EQ. While on the way I modified a few things on these channelstrips and when finished took them to a friend to check them out and get another opinion. This guy is a bigname mastering  and multiplatinum mix engineer here in Germany. It took him 30 seconds of testing to order eight of these channels from me. We tweaked them together to his taste and now he uses them for stem mastering and the occasional mixing project he gets. So much about the quality of the circuit.

I don´t think that it´s a good idea for a hobby-DIYer to second guess the design decisions of a senior circuit designer. Those guys took a lot more into account than we can during their design and evaluation process. Even worse is to transfer single components from one circuit designed for a certain chip to another circuit designed for another chip. That calls for trouble.

If you want to understand things on a component level then you´d better get a good book about opamp circuit basics. Look at the opamp manufacturer´s homepages. They all have free downloads like this: http://web.mit.edu/6.101/www/reference/op_amps_everyone.pdf

Don´t missundestand me, I don´t want to discourage you. But it seems to me that you´re heading into dangerous territory.

 

[Potato Cakes]

Don´t missundestand me, I don´t want to discourage you. But it seems to me that you´re heading into dangerous territory.

I'm not discouraged at all and I'm relying on the knowledgeable members here to avoid catastrophic failure. I have been at this DIY thing for a while and have built projects for clients and use my own creations on all my studio projects and live shows I mix. I'm only now getting to the point where I am starting to see recognizable basic circuits in different designs. So when I started comparing this EQ to others that's when started I wondering about what could be borrowed when tweaking this design.

My goal for the channel EQ would to possibly lower distortion and/or noise. The characteristic that is prevalent in the other EQs I've built is the high mid band can be boosted and there isn't any graininess imparted to the signal. I'd like to have that same characteristic applied here if possible, but I'm not sure where to start. Maybe newer TL072 chips or even NE5532s. I'm pretty sure all the components are the same when this was made new, so maybe in addition to new caps some of the ICs being swapped with updated versions would be beneficial.

And also, I'm using this time on tour for gathering information to aid in making good decisions to prepare a couple of options to do listening tests when the time comes to fire up the iron. So any and all help and guidance is gladly welcomed.

Thanks!

Paul

 

[Potato Cakes]

The PSU is not a problem. Amek did use the same PSU for much larger desks, so don´t worry about that. I had my Scorpion stuffed with current hungry chips and the PSU worked flawlessly, in fact it still works upto today. The only thing I did to the PSU was recapping, adding bypasscaps, renewing the thermal grease and adding a silent fan.
If you draw high current out of any PSU the problem will be the heat. Adding a fan will take care of that.

2x I had other TAC PSUs on the bench with defective fuse holders. In that case it make sense to install new fuse holders with higher current rating. I did use 20A types as replacement.

How large of a value did you use for the PSU bypass caps?

Thanks!

Paul

 

[john12ax7]

I can answer some of the questions when I have some more time.  But I think a good learning step would be to put parts of the circuit in a simulator. Then can change values and get good first hand experience of what things too.

For general things it's important to remember that commercial products are built to a price point.  So most likely you can lower noise and distortion by simply upgrading components to better performing (usually more expensive) ones.

 

[jensenmann]

Usually I use 100nF ceramic caps as PSU bypass caps and Wima MKS2 0,1uF bypasscaps in the signalpath. Ceramics don´t sound good if they are in the audiopath. The exact values are not that critical, anything from 10nF-100nF is ok.

Lowering noise is pretty easy. I did replace all TL072 in the EQ with either OPA2132/4 or TLE2072. Don´t forget to add local bypass caps from B+ and B- to ground as close as possible to each chip you´ve replaced. If space permits then use 22uF electrolytics parallel to those local bypass caps. The selfnoise and distortion figures of these chips are significantly better than TL072s. Along with that modification goes an improvement of the output drive capabilities for the insert point, because the new chips can happily drive 600 Ohm.

Furthermore I did use a BJT input chip as fader buffer instead of the original TL072 FET input chip. The OSI (optimum source impedance) of the TL072 is not so good in that spot, hence increased noise. IIRC I did a LT1358, but LM4562 or LME49720 are fine, too.

To repeat it again: the improvement depends a lot from the local supply bypass caps at each chip. They are necessary. Make it possible to not only add the ceramics but also the electrolytic bypass caps.

 

[john12ax7]

You want R58 = R49 so you have a unity gain inverting amplifier

R41 and R45 II R46 applies positive feedback, the ratio is important. Parallel can allow for non-standard values.

The feedback capacitor you want something that works well at high frequency. Polystyrene is fine but less common these days. Ceramic NPO / COG is a good modern choice.

The cut / boost pot forms a ratio, you should have a flat response at 50% regardless of value. R44 and R47 help to limit the cut / boost range

R48 and R39 form a voltage divider ratio, C30 adjusts this at high frequency, not necessarily things to add to a pre-existing design.

 

[Noisy1]

I like where this thread is going, fun stuff! I rebuilt my Scorpion II about three years ago... with mods still ongoing.

Start the whole process with a full PSU rebuild. The power supply rebuild really opened up the board. The board was really very tired, but a full recap did wonders. AML carries the five-pin caps which made direct replacement easy. I ended up adding a bit more capacitance and adding bypass caps, but that was all well after the basic rebuild was done.  I changed out the fuse holders (as did Jens) as they were an obvious weak point. Added a fan and a switch. The fan makes a huge (HUGE) difference in the cooling of the PSU and is a must-do, not at all optional. I also raised the heat sinks up off of the PCB. Some of the transistor mounts were broken, cracked in half... so some modifications were needed to arrive at a more durable, fail-proof design. I can confidently say that there is nothing wrong design-wise with the AMEK-TAC PSU. I am running 24 channels, 16 busses and the master buss. All is well. Clean, quiet and humming along nicely. Nowhere near the load capacity of the supply.

Without doing the PSU first it is really impossible to make any judgement about the mods you might do. I went down this rabbit hole the other way and it was a mistake.

I put in 24 of the Langley mic mods, and 8 of the Langley line mods. They are good mods, well documented and incredibly easy. They sound excellent. At less than $40 US per channel they are way far on the favorable end of the price to performance ratio. Know that they come with all the caps needed to recap the whole channel, as well as the pots needed for the mod. I am very happy with where it has gotten this board to.

I used a modded Symetrix 202 a lot back in the late 90s and that mod is a pretty smart idea for this console. I have some extra channels so maybe I'll chase after a few channels and make the changes. The Symetrix was a big jump up from the Mackies that were prolific during that time.

I spent a lot of time on the master buss. Every minute of that was well worth the effort. Removing caps, re-chipping, cap selection, better pots...  Most of that stuff is easily searchable here and elsewhere.

I like the EQs on the console. I don't EQ a lot, but they do a great mid-range guitar crunch when called on. I haven't chased after any EQ changes yet, but it is the next obvious step.

On an editorial note, I have maybe $1300 into this console. In a world where $1300 in parts for a compressor build is hardly out of the ordinary, the TAC has been an incredible bargain. I can't imagine anyone will ever be able to see past the reputation of a TAC Scorpion to give me that amount of money for it. Quite likely they will be scared of the "mods" and that will cause even further difficulties in ever selling the console. But for $1300 I have a great console for my personal use, and lower budget pet projects, that does everything way better than I need it to at home.  Its "weird" split console work-flow works for me. I like fiddling with it when I am not building other stuff.  However, I also really like that it works... and that gives me time to work on better tube projects and make music. Frankly it suits its role well. All that said, we still bought a better "world-class" console for the studio for pennies on the dollar of what it was new.

 

[Potato Cakes]

Usually I use 100nF ceramic caps as PSU bypass caps and Wima MKS2 0,1uF bypasscaps in the signalpath. Ceramics don´t sound good if they are in the audiopath. The exact values are not that critical, anything from 10nF-100nF is ok.

Lowering noise is pretty easy. I did replace all TL072 in the EQ with either OPA2132/4 or TLE2072. Don´t forget to add local bypass caps from B+ and B- to ground as close as possible to each chip you´ve replaced. If space permits then use 22uF electrolytics parallel to those local bypass caps. The selfnoise and distortion figures of these chips are significantly better than TL072s. Along with that modification goes an improvement of the output drive capabilities for the insert point, because the new chips can happily drive 600 Ohm.

Furthermore I did use a BJT input chip as fader buffer instead of the original TL072 FET input chip. The OSI (optimum source impedance) of the TL072 is not so good in that spot, hence increased noise. IIRC I did a LT1358, but LM4562 or LME49720 are fine, too.

To repeat it again: the improvement depends a lot from the local supply bypass caps at each chip. They are necessary. Make it possible to not only add the ceramics but also the electrolytic bypass caps.

Just to be perfectly clear, for each chip dual op amp (TL072 replacements)

- 100nF ceramic caps from V+ to ground and V- to ground
- WIMA 100nF (approximately) between A+ to ground, A- to ground, B+ to ground, B- to ground
- 22uF electrolytic parallel with the WIMA caps

The ceramic and film caps I believe I can fit on the underside of the PCB and not interfere with the adjacent channel, but I think I may have to get creative to fit additional electrolytics.

I just finished building 24 of the DOA adapters from Eli-Audio, which I am going to initially use with the new garAM10s. I need to order replacement caps and was wondering if there was any benefit to replacing the polystyrene ones or to just focus on the electrolytics.

While I'm home for a couple of days I'm going to open up the PSU and figure out what value caps needs to be ordered.

Thanks!

Paul

 

[Potato Cakes]

Does anybody have a mouser or digikey link for insulating pads/sheets to use between the heatsink and PCB for the PSU?

Thanks!

Paul

 

[jensenmann]

You can leave the Polystyrene caps in there.
Those Wima film caps are meant to be parallel to each electrolytic in the audiopath (coupling caps). The ceramics are meant to be from rails to ground closet possible to the chips. On top of that (if space permits) use 22uF electrolytics from rails to ground for each chip, as close as possible.

 

[Potato Cakes]

I've been away from this project for quite some time and I'm about a month out from having enough time off to start going really bananas, or just go crazy in general. Hopefully I'll start posting more updates.

In the meantime, I need to order some replacement faders. I believe a good 100mm fader with 120mm mounting holes will fit, but I'm not entirely sure on the dimensions and I'm trying to figure this out while I'm still traveling. Can anyone verify this information?

Thanks!

Paul

 

[shabtek]

Where do you source faders?

 

[jensenmann]

http://www.audiomaintenance.com/acatalog/faders.html

 

[shabtek]

Those 3000 p&g are too dear...

I've read somewhere that p&g are no longer producing faders--so stock is limited.

Anybody try the TDK faders? Any idea on price or vendors?

 

[Potato Cakes]

I've found some Alps and a couple of people with used 100mm 10k faders with 120mm spaced mounting holes.

I'm home now so I'll be able to make the measurements and decide what to do.

Thanks!

Paul

 

[jensenmann]

Anybody try the TDK faders? Any idea on price or vendors?

My studio desk uses these. They are as good as P&Gs. Smooth as butter.

 

[Potato Cakes]

Does any know what the lead spacing of the caps in the PSU are? I'm on the road and I'd like to order parts for the PSU while I am away as there is going to be a soldering/remodeling frenzy taking place when I get home and I'd like to have the parts there when I arrive.

Thanks!

Paul