Op amp replacement, tantalum caps and slew rates

[ebartlet]

Here are some pictures the project:

Channel board

Master/buss board


You can view the schematic for the channel here:

http://www.ericbartlett.com/images/tpchannel.GIF

You can view the schematic for the output section here:

http://www.ericbartlett.com/images/output.JPG

Today I noticed that there are LM301's in the OP section. IC's with a 3.3pf cap to pin 8 are 301's. What should these be replaced with?

Also, Buttachunk, more tants in the OP section?

Thanks

 

[ggoerss]

Replace them with 5534 or similar however remove the capacitor which is used to feed forward compensate the lm301 to improve its slew rate.I did my C12 that way with 5532s in the input pre tlo72s or similar in the eq section and 5534s in the output sections. I used machined pin sockets throughout and also recapped the console in relevant areas.One problem with the C12 is the us of gain controls including faders in the feedback loop of the amplifier.This was done ostensibly to maximise the headroom of each stage however i found that it introduces dc offset into the circuit that causes the faders to be noisy when you move them.

 

[ebartlet]

[quote author="ggoerss"]One problem with the C12 is the us of gain controls including faders in the feedback loop of the amplifier.This was done ostensibly to maximise the headroom of each stage however i found that it introduces dc offset into the circuit that causes the faders to be noisy when you move them.[/quote]

Is there an easy fix for this?

 

[NewYorkDave]

Regarding IC sockets: stray capacitance renders them nearly useless at RF, but the only real problem at audio frequencies is intermittent connections due to oxidized contacts. Using good quality machined pin sockets, as suggested in an earlier message, is the way to go. Also, it doesn't hurt to put a tiny, tiny drop of DeOxit in each socket hole before inserting the IC. Wipe off any excess, of course. This is a kind of overkill, but it will give you some peace of mind if you're really worried about the contacts becoming poor years from now.

 

[ebartlet]

[quote author="buttachunk"]just took another look at your schemos,


not sure if anyone's mentioned this yet, but replace C3 (the one connected to R6) with higher value (1000uf-4700uf 6.3v), and you can replace C5, C6, and C10 with same value.

if you want tantalum in there, go for C1 (does that say 10uf?), and C11 / C18 (22uf/10v)[/quote]

So the moral of the story is replace all signal electros with 1000-4700uf all the way through the stereo buss?

Any other key locations for tants after the channels?

 

[ebartlet]

[quote author="ggoerss"]Replace them with 5534 or similar however remove the capacitor which is used to feed forward compensate the lm301 to improve its slew rate.[/quote]

The one that goes from pin 1 - 2 ?

 

[PRR]

Dude, you're giving me flashbacks. Dig that crazy split-pole compensation on U8. The TIS97 in the clip-lite. And the 2-way pre/post AUX send!!! (gigglefitz)

This is a VERY old IC design. All that crap about replacing TL072s does not apply. Few people today have been (un?)fortunate enough to work on something this old.

> Would the tants in the EQ circuit have any effect on the sound with the EQ at 0?

In this circuit: only the 22/10 caps have any effect with the tone pots centered.

> IC's with a 3.3pf cap to pin 8 are 301's. What should these be replaced with?

301 with a 3pF compensation cap is a VERY good amplifier. Some of these stages also have a 68pF feedforward cap. These are HOT amps, fabulous gain-bandwidth. Slew out the wazoo: more than +/-18V audio ever needs. Unlike lesser fixed-comp amps, you may not get any great improvement from replacing these.

The only thing I'd suggest: use sockets (they do not give trouble) and a dozen 301s. Start swapping chips in one socket, and listen to the soft passages. About 1 in 10 chips may sound "off" or even "fuzzy": crossover distortion is not well controlled on 301 type chips. Maybe 1 in 100 will have high hiss. Weed-out the bad ones, and re-chip everything.

Also: leave the 0.01uFd ceramic caps as power-pin bypass, but supplement them with a 0.22uFd poly and a 1,000uFd 25V aluminum for every 2 or 3 chips, to suck-up power-rail garbage. (Feedforward 301 is very susceptable to power crap.)

What are the other mystery chips? Particularly the mike-amp?

 

[ebartlet]

[quote author="PRR"]301 with a 3pF compensation cap is a VERY good amplifier. Some of these stages also have a 68pF feedforward cap. These are HOT amps, fabulous gain-bandwidth. Slew out the wazoo: more than +/-18V audio ever needs. Unlike lesser fixed-comp amps, you may not get any great improvement from replacing these. [/quote]

Thats suprising, I have only heard negative things about the 301.

Hmmm... keep the 301's or put in 5534's...

[quote author="PRR"]Some of these stages also have a 68pF feedforward cap.[/quote]

If I decided to keep the 301's, should I replace the ceramic 68pf? polyprop?



[quote author="PRR"]What are the other mystery chips? Particularly the mike-amp?[/quote]

Everything else is 4558's.

 

[ebartlet]

[quote author="buttachunk"][quote author="ebartlet"]So the moral of the story is replace all signal electros with 1000-4700uf all the way through the stereo buss?[/quote]


eh ?
no way--- just that one pointed out, C3 !

replace all others with same value (if you are going to recap).[/quote]

[quote author="buttachunk"]
just took another look at your schemos,


not sure if anyone's mentioned this yet, but replace C3 (the one connected to R6) with higher value (1000uf-4700uf 6.3v), and you can replace C5, C6, and C10 with same value. [/quote]




Ah, I realized that I misunderstood you. You meant same value as original opposed to the same as C3.

That makes more sense to me and my limited understanding...

 

[ebartlet]

I'm considering BC's for coupling caps. Any thoughts?

 

[microx]

Just one observation, I would do the mods in stages, like say 4558's for 5532's then see how you like it then some more. That way you will know what changed what. On the old Yamaha mixer I am working on just now (which is also FULL of 4558's) I worked on the PSU first as this will always be beneficial and then I started on 5532's for the 4558's. Also I left one channel untouched as a reference to go back to. Use sockets, the benefits way outweigh any drawbacks.
Good luck, these kind of rebuilds are very rewarding,
Steve

 

[PRR]

> Everything else is 4558's

That may be a problem. Lose them. At the least, use one of JRC's improved 4558s; the 4559 may be suitable. TL072 will also work OK, lower THD though a little more noise. Off hand I don't see any place a 5532 won't work.

What have you heard of 301s? They were not used for long, but were used in a few very fine systems, when properly compensated.

 

[ebartlet]

[quote author="PRR"]>

What have you heard of 301s? They were not used for long, but were used in a few very fine systems, when properly compensated.[/quote]

That its old and slow (slew rate), and that 5534's are better...

whatever that means...

Of course thats what you get on the internet...

I was supprised at the slew rate specs for the 301, 10V/us!

Yes the slew is higher for the 5534, but not much.

I think I'll leve them alone for now, unless I get some really good reasons to change em.

l be busy enough swapping caps and chips in the channels

Should I replace the ceramic feed forward cap with something else?

 

[StephenGiles]

But how does slew rate change the sound? Does it change the sound at all?
Stephen

 

[ebartlet]

[quote author="StephenGiles"]But how does slew rate change the sound? Does it change the sound at all?
Stephen[/quote]

Too low of a slew rate, causes loss of detail of high frequencies/transients because the voltage change happens faster than the amplifier reproduce. The fast rise gets rounded.

With modern opamps I don't think slew rate is as much of an issue (especially to untrained ears) as the other specifications.

For example the opamps in a mackie have a slew rate of 4V/us and I have heard that some API stuff exhibits 3V/us. Most of us would agree API stuff sounds better. OTOH, A John Hardy 990 will do 18V/us and sounds more detailed (cleaner) than the API

At higher slew rates it becomes more difficult to tell the difference.

 

[StephenGiles]

Thanks for that ebartlet, I've just done a before and after test on my BEHRINGER mixer and am relieved that it must use high slew rate opamps.
Stephen

 

[StephenGiles]

But 4458s sound great in a Rat distortion clone!
Stephen

 

[ebartlet]

[quote author="StephenGiles"]But 4458s sound great in a Rat distortion clone!
Stephen[/quote]

Great! Anybody want to buy some slightly used vintage 4558's I'll be pulling out of my board?

They are vintage so they must be worth alot, right? :wink:

 

[Gus]

It is not all slew rate as your ears tell you. Alot of an opamps sound has to do with the input pair. Most opamps do not have emitter Rs in the input diff pair. Then you get to the second stage intergrater for your stablity then the output section.

I find I like opamps with alot of current output abilty. This also seems to helps "control" the input stages nolinearitys by having more current to correct the input via feedback.

High current out with low value feedback Rs helps alot.

Now if TL07x had a good output stage then might be great I can't stand them if they need to drive anything lower than a 10K load

The 318 is great except for the noise because of the input degeneration.

What do the big opamps have? big current ,more offset problems because of higher current in the input stage if BJT some even have emitter Rs or inductors!

So many things that don't fit together well

 

[ebartlet]

[quote author="Gus"] What do the big opamps have? big current ,more offset problems because of higher current in the input stage if BJT some even have emitter Rs or inductors!

So many things that don't fit together well[/quote]

It's all making more sense to me. Did I hear correctly that the API stuff can drive a 75 ohm load with no problems?