The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Gear for sale  Search Gearslutz Go Advanced
symetrix 525 compressor mods?
Old 11th September 2006
  #1
Gear Addict
 

Thread Starter
symetrix 525 compressor mods?

I just acquired 2 of these things at a very decent price. They sound ok, but it seems as if somethings holding it back from shining like it could.

I don't know anything about analyzing circuits, so I was wondering if anyone could glance at the schemo's and see if anything stands out that could be done different or better?

symetrix 525 schematics


If so, these can be had for around alesis 3630 money, and they used REAL Neutriks in back!!
Old 11th September 2006
  #2
Lives for gear
 

Hi
Is it the overall 'tone' of the unit, when it is not compressing or expanding, or is it the 'dynamics' that you feel could do with a boost?
There are better chips than the RC4560 available now and it may do with a sprinkling of new caps, possibly mylar bypasses as well. This may put a little sparkle to the audio chain (sheet 1 of schematic). Possibly best leaving the sidechain alone as you could spend a while unravelling it.
Be careful what you use for the U2 (102) positions as it is part used as a comparator and not all chips like being used in this way. U1 (101) should have np problems.
Matt S
Old 11th September 2006
  #3
Lives for gear
 

$.02

beyond basic idea of using better parts what is it you dont like about it ...why does it need to change???? with out specific data cant really suggest in generic terms....sorry
Old 11th September 2006
  #4
Gear Addict
 

Thread Starter
Well there's nothing really WRONG with it now, and i know i shouldn't be comparing a 60.00 unit to my ssl G384, BUT:

I guess i was just wondering if there was anything really obvious, like cap values or chip changes that can open it up, make it a little faster sounding etc...
Old 11th September 2006
  #5
Lives for gear
 

I don't know what the 'sound' of the Alison gain element is, compared to the DBX VCA chips but otherwise the Symetrix COULD be cleaner as there is less 'stuff' in there, especially if you swap the op amps. The SSL uses 5534's or 5532's but do not immediately stick a 5532 into U2 position as part is used as a comparator (see above). You need a chip which does not get unhappy if you wrench it's input 'legs' apart (electrically). 5532 would probably not blow up but the indicator may not indicate as it should.
Otherwise go for fresh caps as I presume it is getting on a bit.
Matt S
Old 24th June 2015
  #6
Lives for gear
 
Silent Sound's Avatar
I'm reviving this old thread because I recently picked up a 525 and like how it works, but hate how it sounds. Specifically the noise and lack of bandwidth (both at the bottom and top of the frequency spectrum). So I'd like to modify it to address the noise and bandwidth issues (mostly noise).

http://www.symetrixaudio.com/kb/525_sch.pdf

First up is the power supply (page 3 of the schematic). I'm going to upgrade the rectifier diodes and regulators with the same type, only higher quality (1N4003GP, LM7815ACT, LM7915CT). I'm also going to replace that string of .01uF caps (C205-C218) with .1uF mono ceramic caps. How about the two big filter caps? I'm thinking of replacing the 4.7uF (C203, C204) with 47uF caps and don't know about the 1000uF (C201, C202). Maybe go with 6800uF? I'm afraid to bump that one up too much. I'm more knowledgeable with tube amps and know that going with too large of a filter cap can stress the rectifiers and PT. Solid state is a bit of a mystery to me. What do you guys think?

After that, I'm just gonna do the usual stuff to clean up the audio signal path (page 1 of the schematic). Basically just replace all of the resistors with metal film ones and replace all of the capacitors with film or mica ones of the same value. My main question here is the two opamps and the large electrolytic cap at the output. I'm guessing there's probably something better than a RC4560 by now. Maybe an LM6172? How about the new OPA2209? Try them both and find out? Once again, tubes I can do. IC chips, I don't know anything about. And what about C10, that large 100uF electrolytic cap at the output? Any ideas for a good replacement with that, or just swap it out for a higher quality electrolytic of the same value?

The sidechain I plan to leave as is. It works well enough as is. Does all of this sound reasonably solid, or am I about to make an overly complex, one time use smoke machine?
Attached Files
File Type: pdf 525_manual.pdf (1.92 MB, 129 views)
Old 24th June 2015
  #7
Lives for gear
 

Hi
Most of the 'plans' are unnecessary and even if you change the op amps you may well find the EGC101 is the noise limiting factor.
Unless you are suffering 'hum' issues (ripple from the supply) then there is not a lot of point making the 1000uF caps much bigger than say 4,700uF and even 2,200 would be reasonable.
As the power transformer will be fairly small the inrush current for BIG capacitors should be OK as the 1N4000 series are OK for single peaks of 50 Amps.
As is usual in any of these situations FIND OUT WHERE the noise is coming from. There is no point 'upgrading' the supply if it is not causing the problem.
Unless it is built using carbon resistors (unlikely) there is no point in swapping those.
Similarly unless you can 'prove' the regulators are actually contributing noise to the output, there is no benefit to change.
Matt S
Old 24th June 2015
  #8
Lives for gear
 
Silent Sound's Avatar
Quote:
Originally Posted by Matt Syson View Post
Hi
Most of the 'plans' are unnecessary and even if you change the op amps you may well find the EGC101 is the noise limiting factor.
Unless you are suffering 'hum' issues (ripple from the supply) then there is not a lot of point making the 1000uF caps much bigger than say 4,700uF and even 2,200 would be reasonable.
As the power transformer will be fairly small the inrush current for BIG capacitors should be OK as the 1N4000 series are OK for single peaks of 50 Amps.
As is usual in any of these situations FIND OUT WHERE the noise is coming from. There is no point 'upgrading' the supply if it is not causing the problem.
Unless it is built using carbon resistors (unlikely) there is no point in swapping those.
Similarly unless you can 'prove' the regulators are actually contributing noise to the output, there is no benefit to change.
Matt S
Thanks for the reply. I'm hoping to make the EGC101's the limiting factor. I guess I should scope out the path and see what I can find, noise wise. Still, with so many ceramic caps in the signal path, I'd have to think there would be some benefit to a general upgrade of components. The resistors are carbon film. So while metal film aren't really much of an improvement, I've got those parts on hand anyway, so I figured why not? I'll save them for last and see how long the other stuff takes and make my decision on what to do with them when the time comes. I may just leave them if I start losing interest. And that's half of why I'm doing this. Part of it is to get a usable clean hardware comp. to use as a safety net to tamp down any wild peaks while tracking highly dynamic sources. So it wouldn't be used often, but it would be nice to have something available for that rare circumstance. Secondly, I just like doing projects!

In any case, this thing is really noisy! Even with nothing plugged in and the thing on bypass, it still hisses like an old cassette player. I can't say I've heard much hum, but it could be masked by the noise. Maybe the scope will tell me more. Part of the power supply upgrade idea is to quiet down the device. The other part is to make sure I get enough clean power to the opamps to try other, more power hungry IC's and see what works. I'm thinking at the very least I should swap out the electrolytics, as this thing is close to 30 years old. So maybe I'll start off with those and then scope it and see where I stand. I got this thing dirt cheap, so I'm feeling pretty fearless with it.
Old 24th June 2015
  #9
Lives for gear
 

Hi
Scaling down the resistors in the input diff amp (say factor of 2) will reduce some of the noise, and fitting quieter op amps, checking they are suitably 'bypassed' with 0u1 caps very close.
The 13 K's around the VCA should not be reduced as it will increase distortion.
Check that the noise is around -86dBu (measured with a RMS meter with 20 - 20KHz bandwidth). Using a 'soundcard' or DAW will not necessarily give you the correct bandwidth and 'characteristics' for meaningful measurements.
Matt S
Old 24th June 2015
  #10
Lives for gear
 
Silent Sound's Avatar
Quote:
Originally Posted by Matt Syson View Post
Hi
Scaling down the resistors in the input diff amp (say factor of 2) will reduce some of the noise, and fitting quieter op amps, checking they are suitably 'bypassed' with 0u1 caps very close.
The 13 K's around the VCA should not be reduced as it will increase distortion.
Check that the noise is around -86dBu (measured with a RMS meter with 20 - 20KHz bandwidth). Using a 'soundcard' or DAW will not necessarily give you the correct bandwidth and 'characteristics' for meaningful measurements.
Matt S
Thanks Matt! That will give me enough to get started. I'll try this and see how it goes and post some before and after clips.
Old 24th June 2015
  #11
Lives for gear
 
Silent Sound's Avatar
I'm an idiot. I had the passive summing mixer connected to the compressor and then the preamp instead of the mixer going into the preamp and then the compressor. No wonder the noise was hideous! It's much quieter now, of course. It still seems to dull things a bit and has a little bit of noise, so there might be some room for improvement, we'll see. Some of that is the nature of the compressor, I'm sure. Like I said, it's as much a fun project to work on as anything.
Old 30th June 2015
  #12
Lives for gear
 
Silent Sound's Avatar
Okay, so here's what I wound up doing:
Main filter caps 1000uF-2200Uf
Upgraded all .01uF filter caps to .1uF mono ceramics
Upgraded all feedback bypass caps in signal path with silver mica (previously they were all ceramic 0.47uF. I switched them to the values on the schematic)
Upgraded all 4.7uF nonpolar E-lytics with giant ass Wima MK4 caps that I had to attach flying leads to and hot glue down (should have paid more attention there).
Upgraded 100uF output E-lytics to audio grade low ESR E-lytics
Upgraded RC4560 IC's in signal path with LM6172's.

As promised, here are the before and after clips when used on the master bus (24/44.1). I didn't go a great job of level matching them so the post mod clip is a bit louder. To be fair, the unit seemed a bit louder, post mod. I tried to keep the settings the same in both clips, but there's probably some variation there. I used a camera to record and line up the knobs. I had gate and ratio maxed out and threshold around +15 and the gain around 2.

So my goals were to lower the noise floor and increase the frequency response. To me, it sounds like the noise floor has definitely been lowered. As for the frequency response, it seems like it might have improved slightly, especially in the high end.

Thanks for you help Matt!
Attached Files

Post Mod.wav (7.45 MB, 1045 views)

Pre Mod.wav (7.45 MB, 1037 views)

Old 4th July 2015
  #13
Lives for gear
 

The limited frequency response in the high end (- 1db at 20k according to the specs) might have something to do with the fairly big caps in the NFB path. They're very probably that big for a reason, too.

Quote:
it seems like it might have improved slightly
Did it or didn't it? This is easy to check even with a basic sound card, there should be no doubt about it.


Have you checked for stability? The LM6172 is a very nice op amp, but it's also a very fast part and might oscillate in a circuit designed for a much slower op amp.
Old 4th July 2015
  #14
Lives for gear
 
Silent Sound's Avatar
I probably should have done some frequency sweeps and a better test than running a full mix through a summing mixer and a preamp without detentes, but I wasn't interested in getting that scientific with it. I was thinking along the lines of "if I can't hear the difference, then there's not a significant difference". It sounds a bit clearer up top, but with the variables in all of the knobs involved it's hard to be certain. So the difference in high frequency, if anything, is small. The big difference, which was clearly noticeable, was in the noise floor.

As for stability, that's hard to check. It looks great on my scope, but it only goes up to 100 MHz. The chips don't get very hot or have any more DC offset than the other RC4560D chips in the sidechain. So I'm just going to not worry about it and if it fails, it fails. I'm not very deep into it financially, so it wouldn't be worth it to me to buy more equipment to investigate further.
Old 4th July 2015
  #15
Lives for gear
 

Hi
Ci and 2 (4u7) could be linked out as long as C7 is a poly??? type (very low leakage). Could take C7 to 10uF maybe?
The 2, 470pF caps could go to 220 or 270pF and the other 2 caps on the FIRST op amp could reduce by say half (keep the ratio roughly equal).
This will increase the HF a bit but MIGHT let a bit of RF in, depending on your studio setup.
You could tinker with the other 2 feedback caps but these might get you closer to trouble so more 'experimental'.
Matt S
Old 4th July 2015
  #16
Lives for gear
 

Had not seen the schematic. There are no caps in a feedback path higher than 150p max, .047uf would have been excessive.
Old 4th July 2015
  #17
Lives for gear
 

Hi
I would actually suggest replace C1 and 2 with a ferrite bead (thus DC coupled). C7 then does the job of removing all offset before the VCA itself.
150pF is quite 'large' if the feedback resistors are 47K region.
Matt S
Old 4th July 2015
  #18
Lives for gear
 

Quote:
Originally Posted by Matt Syson View Post
Hi
I would actually suggest replace C1 and 2 with a ferrite bead (thus DC coupled). C7 then does the job of removing all offset before the VCA itself.
But that would turn a high pass filter into a low pass filter, wouldn't it? Then you would probably no longer need C3 and C4, but U1 won't like it if someone accidentally input phantom power.
Old 4th July 2015
  #19
Lives for gear
 

Hi
You want a low pass filter to keep 'RF' out, which is what the 470pF caps with the 2K0 resistors are doing. Having a ferrite and lowering the cap value 'kind of' balance each other out in that the cap becomes more effective but the response stays 'flatter' at lower frequencies. (Badly explained!)>
Yes you would lose the 'phantom power protection' but the resistors would limit current anyway and it is rare for all gear to have EFFECTIVE phantom power 'protection'. The protection of a 'cap' would not actually protect from it being plugged or switched (it would 'spike') but it would reduce long term DC. The 'spike' would do just as much damage most likely.
Matt S
Old 4th July 2015
  #20
Lives for gear
 

Quote:
Originally Posted by Matt Syson View Post
Hi
You want a low pass filter to keep 'RF' out, which is what the 470pF caps with the 2K0 resistors are doing. Having a ferrite and lowering the cap value 'kind of' balance each other out in that the cap becomes more effective but the response stays 'flatter' at lower frequencies. (Badly explained!)>
Yes you would lose the 'phantom power protection' but the resistors would limit current anyway and it is rare for all gear to have EFFECTIVE phantom power 'protection'. The protection of a 'cap' would not actually protect from it being plugged or switched (it would 'spike') but it would reduce long term DC. The 'spike' would do just as much damage most likely.
Matt S
Sure, filtering out RF is what the lowpass is about and what a ferrite bead would do. But the low end frequency / phase response with only the bead would not just be flatter but "flat", since there is no high pass filter before U1A. A ferrite-based high pass (RL) would require the beads to be inserted in parallel.

You're right about the spike, I guess.
Old 4th July 2015
  #21
Lives for gear
 

Hi
[ A ferrite-based high pass (RL) would require the beads to be inserted in parallel. ]
Yes but we want LOW pass.There should be minimal 'DC' on incoming signals and any other 'working' gear would have little or no DC on it.
We DO want LOW pass to remove RF picked up on cables that MIGHT get demodulated by the op amp.
Thus ferrites in SERIES, which can conveniently be fitted where those 2 4u7 caps were.
If the original circuit was deliberately using 4u7 as 'low pass filtering' with ANY intrusion into the wanted audio signal it would be a very poor design choice. Not least in the fact that they should be 'matched' to give common mode rejection (balanced).
Matt S
Old 8th July 2015
  #22
Lives for gear
 

Somehow didn't get an email notice about your reply.

The circuit designers also thought there should be a high pass filter, thus C3 and C4 which block RF.

But what function would C1 and C2 have other than blocking DC?

Quote:
Originally Posted by Matt Syson View Post
Hi
[ A ferrite-based high pass (RL) would require the beads to be inserted in parallel. ]
Yes but we want LOW pass.There should be minimal 'DC' on incoming signals and any other 'working' gear would have little or no DC on it.
We DO want LOW pass to remove RF picked up on cables that MIGHT get demodulated by the op amp.
Thus ferrites in SERIES, which can conveniently be fitted where those 2 4u7 caps were.
If the original circuit was deliberately using 4u7 as 'low pass filtering' with ANY intrusion into the wanted audio signal it would be a very poor design choice. Not least in the fact that they should be 'matched' to give common mode rejection (balanced).
Matt S
Old 8th July 2015
  #23
Lives for gear
 

Hi
It is conventional to block DC on inputs however it is usually minimal and the stage only has unity gain and it's output is not switched (during usual operation) so a few millivolts of DC wouldn't hurt the audio. Therefore C1 and C2 could go.
Full protection against phantom power or massive 'spikes' or silliness is not provided anyway.
Matt S
Old 9th July 2015
  #24
Lives for gear
 
ulysses's Avatar
You don't need protection against phantom power on the input of a line-level device. Phantom power appears on preamp *inputs* so you only have to worry about it accidentally being fed into another device's *outputs*. This of course assumes you don't do something silly like put your mike preamp inputs on the TRS patchbay where they could be patched to any input or output.

Personally, I would use low-offset op amps and then replace C1, C2, and C10 with jumper wires.
I would use NPO ceramics for the feedback caps and see if I could lower C9 a bit (47pF?) without losing stability. But first I would measure the bandwidth of the whole circuit.
Old 9th July 2015
  #25
Lives for gear
 

Hi
Most op amp inputs have reverse 'catching' diodes to protect against excess voltage on inputs. These are rated at 10mA or thereabouts.
With 47K input resistors even 240 Volts DC would only put about 5 milliamps into the input.
Ferrite beads rather than links may help keep your cellphone out of the mix.
Matt S
Old 11th August 2015
  #26
Lives for gear
 
antichef's Avatar
Great info in this thread - I'm definitely cracking open my crumbly 525. Thanks for pointing me over here Silent Sound!

I get a kick out of geeking out this way so even if the patient doesn't survive it's not a complete loss.

Do you all have problems with the pots? Mine are in various states of decay and malfunction. I had to cannibalize one from the gate (which I don't use) to keep the compression controls working, and I can't seem to find replacement parts.
Old 12th August 2015
  #27
Lives for gear
 
Silent Sound's Avatar
Quote:
Originally Posted by antichef View Post
Great info in this thread - I'm definitely cracking open my crumbly 525. Thanks for pointing me over here Silent Sound!

I get a kick out of geeking out this way so even if the patient doesn't survive it's not a complete loss.

Do you all have problems with the pots? Mine are in various states of decay and malfunction. I had to cannibalize one from the gate (which I don't use) to keep the compression controls working, and I can't seem to find replacement parts.
The pots on mine work fine. I bet Mouser has those pots. I've bought some crazy pots from them in the past. I did have to wait for a couple of months once. But that was a detented audio taper with an odd shaft size. At the least, they'll have one that could work with some flying leads.
Loading mentioned products ...
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get instant access to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
drunton / Gearslutz Secondhand Gear Classifieds
2
recky / High end
10
chriscoleman / Gearslutz Secondhand Gear Classifieds
2
idylldon / Gearslutz Secondhand Gear Classifieds
0
Saucyjack / High end
14

Forum Jump