Trying to learn about opamps over here.
Looking at this schematic for a mixer I have, it has a 5532 opamp in the U1 first stage and then three 4558 opamps (shared between two channels) for the EQ and effects send in U2 U3 U4 U5.
From what I've read it sounds like 4558 is pretty much bottom of the line for acceptable audio use. Can they be replaced with 5532 in these positions? Is there something else that would actually improve it, in a DIP-8 package? It's a low-end mixer I'd like to improve a couple channels, but not so much that I'd want to bother with soldering caps around to stabilize more sophisticated chips. It sounds okay but pretty dull in the high frequencies and very sloppy in the lows.
Would 5532 work here, and if not, could someone tell me what the necessary parameters for this kind of circuit? Thanks in advance.
Looking at this schematic for a mixer I have, it has a 5532 opamp in the U1 first stage and then three 4558 opamps (shared between two channels) for the EQ and effects send in U2 U3 U4 U5.
From what I've read it sounds like 4558 is pretty much bottom of the line for acceptable audio use. Can they be replaced with 5532 in these positions? Is there something else that would actually improve it, in a DIP-8 package? It's a low-end mixer I'd like to improve a couple channels, but not so much that I'd want to bother with soldering caps around to stabilize more sophisticated chips. It sounds okay but pretty dull in the high frequencies and very sloppy in the lows.
Would 5532 work here, and if not, could someone tell me what the necessary parameters for this kind of circuit? Thanks in advance.
Last edited:
That looks like a Peavey schematic from the 90's. I've rebuilt a few of 'em in the past.
How many channels in the board? An important consideration: How much more current is available from the +15 and -15 VDC supplies? Good to know that before rolling opamps. Almost anything worth using is going to draw more quiescent current than a 4558.
U5 just drives LED's, no need to change it.
U2 and U3 are high impedance circuits and would benefit from being a FET opamp.
U4 is just an inverting buffer, something like an NJM2068 would work fine.
U5 drives the output busses and direct out, I'd use a 5532 there.
The main somic limitation with the old Peavey mixers is the power supply and grounding schemes. They're just too spead out and leggy to give top-notch performance.
The best bang-per-buck upgrade on these is to improve the buss mixing amps.
How many channels in the board? An important consideration: How much more current is available from the +15 and -15 VDC supplies? Good to know that before rolling opamps. Almost anything worth using is going to draw more quiescent current than a 4558.
U5 just drives LED's, no need to change it.
U2 and U3 are high impedance circuits and would benefit from being a FET opamp.
U4 is just an inverting buffer, something like an NJM2068 would work fine.
U5 drives the output busses and direct out, I'd use a 5532 there.
The main somic limitation with the old Peavey mixers is the power supply and grounding schemes. They're just too spead out and leggy to give top-notch performance.
The best bang-per-buck upgrade on these is to improve the buss mixing amps.
Thanks that is very helpful.
The mixer is 12 channels, but I don't need all of them. It would be nice to have 4-6 real nice sounding preamps.
I guess it's the ones that drive the EQ section (U2 U3?) that seem hissy, but I don't need to use a lot of boost there.
If I am using fewer channels/current can I do better than 5532 in U1?
What would you recommend for the output buss amps? Did you mean 5532 there?
The mixer is 12 channels, but I don't need all of them. It would be nice to have 4-6 real nice sounding preamps.
I guess it's the ones that drive the EQ section (U2 U3?) that seem hissy, but I don't need to use a lot of boost there.
If I am using fewer channels/current can I do better than 5532 in U1?
What would you recommend for the output buss amps? Did you mean 5532 there?
The EQ opamps are noisey even without boost. A FET will help.
Maybe an LM4562 for U1. But you might not hear the difference. Like I said, the power supply and grounding scheme limits how much it can be improved.
Sorry, I meant the buss summing amps on the main board. They're doing alot of the work. Something like a 4562 would be good there. But they probably already have a 5532.
Maybe an LM4562 for U1. But you might not hear the difference. Like I said, the power supply and grounding scheme limits how much it can be improved.
Sorry, I meant the buss summing amps on the main board. They're doing alot of the work. Something like a 4562 would be good there. But they probably already have a 5532.
Yes the master buss section has 5532s but also a couple 4558s (maybe those are the monitor buss? will check).
For the EQ amps, would a TL072 be quieter because it's a FET, or can I just use 5532s all around?
If a different amp in U1 doesn't make it really hi-fi for $10-$20, it's probably not worth bothering.
I will definitely look at working on the power supply.
For the EQ amps, would a TL072 be quieter because it's a FET, or can I just use 5532s all around?
If a different amp in U1 doesn't make it really hi-fi for $10-$20, it's probably not worth bothering.
I will definitely look at working on the power supply.
Yes, because it's a FET. But you can do better than a TL072...
Agreed.
Don't mean to be discouraging, but I wouldn't waste too much time on it (been there...). The problem is not the parts used, it's the basic design/layout. A single power supply and circuit ground distributed by a couple skinny buss wires to 12 separate channels, each with long, thin pcb traces for rails and ground to the opamps. It's not a good setup. The real fix is to have regulators on each channel. A real PITA to do.
This is a sound reinforcement mixer, yes?
Yeah dig it. It's just a freebie I got, want to make some good use of it, not high level stuff. It is a live PA type mixer, yes. What's better than a TL072 and less than say $7, if I can't use the 5532?
While we're on the subject, do you have any insight on this? I have been playing around with this opamp noise calc and for a bass amp with LF353s in the preamp... it seems I can't do much better than that LF353 for noise, which is surprising. Not sure how to calculate the non-inverting/inverting resistance of the second half of U1. It's not a noisy amp, but the first gain stage is not that quiet turned up, and the high EQ has the hiss you would expect. No way to reduce that? There are no crappy resistors or caps to replace.
While we're on the subject, do you have any insight on this? I have been playing around with this opamp noise calc and for a bass amp with LF353s in the preamp... it seems I can't do much better than that LF353 for noise, which is surprising. Not sure how to calculate the non-inverting/inverting resistance of the second half of U1. It's not a noisy amp, but the first gain stage is not that quiet turned up, and the high EQ has the hiss you would expect. No way to reduce that? There are no crappy resistors or caps to replace.
OPA1642, but it's SOIC so you need to mount it to a DIP8 adapter. I'd even take an LF353 over a TL072.
With impedances that high, current noise dominates and you're pretty mucvh locked into using FET opamps.
Okay, maybe a LF353 is a good choice then, they're affordable. SOIC is not my idea of a good time.
Glad to know that, thanks, I'm just here to learn about opamps. Looks like an AD745 might save me 2db of noise... in each stage 🤷♂️ Might try it just for kicks. I could have sworn I read a thread where a guy got lower noise in this amp by swapping these opamps, but darned if I can find it now.
Also a lot of the resistors are crap! It’s kind of a never ending process to lower the noise and hiss on stuff like this. If you have any high value carbon comp resistors you can use carbon or metal film in their place. Make sure your power supply has fresh caps including the decoupling caps at the ICs. It’s never going to be dead quiet but it will be better.
I'd vote for the OPA1642 too. Very recent design with good overall behaviour and very good price point. Need to use adapters though...
All these resistors are metal film - I think all electronics since the late 70s would be - so unfortunately no easy fix there. It's a neverending process that never goes anywhere! I went down that rabbit hole once, replacing every resistor in a Fender tube amp (from carbon comp to metal) and seemed to have no effect anyway.
The mixer could use new power supply caps but I won't expect a change in the sound. It's funny how hard it is to improve a device with component changes - and yet this mixer will be nowhere near as good sounding as say, a Sound Devices MixPre3 field recorder, for example, which is $500. Hard to bridge that gap.
As for the 1642 opamps, in this case it's not worth the cost of labor (mine) - it would have to be a DIP-8 package. It does seem like the 4558 in there is the worth option.
Unlikely... certainly not on the Peavey mixers I worked on. Most were carbon film. But I agree with your assessment - you've got bigger fish to fry before that would matter. Like nine small 'lytics in the signal path!
Agreed. Huge inductance between the regulators and the ICs on the channel strips is the biggy. Lots of crosstalk between channels through the rails.
Something else to consider: With such high impedance in the power supply rails and circuit ground at the IC's, you can't count on local decoupling caps being effective.
BTW, the AD745 is s single, the AD746 is the dual. I'd take an OPA1642 over it. But most of the opamps are used in inverting mode so the difference wouldn't be as great anyway.
Last edited:
Oh brother. All those electrolytics are in the signal chain eh? Mixer is probably not worth working on, I think is the verdict.
With the bass amp, it sounds like knowledgeable folks report an appreciable improvement with AD823 or OPA2134, so I am inclined to have a go with that in the U1 and U2 positions. When fixing it I had to replace the 351 on the power amp board and used a TLE2071 instead, and that was a marked improvement.
With the bass amp, it sounds like knowledgeable folks report an appreciable improvement with AD823 or OPA2134, so I am inclined to have a go with that in the U1 and U2 positions. When fixing it I had to replace the 351 on the power amp board and used a TLE2071 instead, and that was a marked improvement.