Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Greg Norman]

I don't know if this topic has been beaten out on other forums, but I was wondering if anyone had any luck replacing those op amps (HS-1000 or similar).  So far I've made a daughter board adapting standard DIP amps with +-20 volt regulation to the footprint of the Sontec amp.  Using a 5534 for a test (for the input amp) and not changing any surrounding components, the eq worked quiet and predictable for 2 days.  This was a crude test but I just wondered if anyone has tread the same path. Ultimately I'm shooting for a fast low noise DIP amp for the input and the Hardy 990 amp for the output amp.  I'm not too concerned about maintaining the marginally higher headroom or emulating the sound of a vintage Sontec MEP-250B.  I just want to make use of this eq.

Any suggestions.
I'll post all that I learn.
Thanks,
Greg

[George Massenburg]

Greg Norman wrote on Fri, 09 July 2004 17:57

I don't know if this topic has been beaten out on other forums, but I was wondering if anyone had any luck replacing those op amps (HS-1000 or similar).  So far I've made a daughter board adapting standard DIP amps with +-20 volt regulation to the footprint of the Sontec amp.  Using a 5534 for a test (for the input amp) and not changing any surrounding components, the eq worked quiet and predictable for 2 days.  This was a crude test but I just wondered if anyone has tread the same path. Ultimately I'm shooting for a fast low noise DIP amp for the input and the Hardy 990 amp for the output amp.  I'm not too concerned about maintaining the marginally higher headroom or emulating the sound of a vintage Sontec MEP-250B.  I just want to make use of this eq. Any suggestions. I'll post all that I learn. Thanks, Greg

You know, Sontec is a tough issue for me.  

After ITI folded, I was a partner in Sontec, and all of the key designs were mine.  For years I felt I had to support Burgess' manufacturing.  But for as many times I brought him solutions to manufacturing problems I found that nothing I said or did would keep him from making really bad choices...and who knows why?  To save money?

After rebuilding several of the more complicated boxes (the Sontec mastering EQ for Kendun comes immediately to mind, where I spent days rebuilding all of the op amps and most of the unit), I decided he'd have to pay me something for the extraordinary support I was putting in.  I got nothing but bullshit (those of you who've spoken to Burgess on the phone know what I mean).

And thus I made what I think is the correct decision to step completely away from supporting any Sontec products.

Fixing one is one of life's most ungratifying jobs, and I wish you luck.  Sorry I don't have any answers for you.  I can tell you this.  You're on the right path, except that there're aren't many chips that can run +/- 28v.  Although the Hardy goes to +24v doesn't it?  Anyway, you could probably do it with all 5534's if you run the Vcc down to +/- 22 volts or so (we've run quite a few of the old Signetics there).  The problem is that the Sontec was laid out for a FET-input op amp, which as you all know has zero input current.  The 5534 and almost all bipolar input chips have significant input current, and the frequency pots on the EQ have to supply that current.  Not to mention the output offset, which will result in a click on the in/out switch.

Anyway, let me know how you are getting along.

George

[bblackwood]

Greg, I have lines on a couple of guys building replacement boards for the HS1000. PM me if you want more info.

Trying the get them from Burgess is impossible - he dropped off the face of the earth a few years ago. I have just enough amps to keep my box running for a while, but soon enough...

And George, I don't think anyone would expect you to support a product you left behind decades ago. I certainly don't.

[Dan Kennedy]

Just as a possibility for a more plug and play part than the 5534, the OPA-604 from Burr-Brown will take +/-24v, it's a FET input, and has pretty fair drive capability.

Digi-Key has them in an 8 pin dip package for $2.40 in low quantities.

[Greg Norman]

Thanks Dan, Brad, and George.  
I've never used the OP604s.  People have said good things about them.  I'll pick up a few in the next order.  
I built a channel with the 990 op amp as the o/p driver (5534 as the i/p amp) and did some listening tests yesterday.  After calibrating the offsets and noiseless switching circuit I couldn't notice a change in sound quality.  I had a mono signal split to the two inputs.  I monitored the o/p as a stereo signal with the eq in bypass (signal goes through the two replacement amps in bypass).    There was no deviation from mono.  I like listening to things this way to see if there is a difference, then trying find out what the difference is (phase shift, eq, level, and so on).  In this crude listening test the two channels sounded almost identical (eq in and out).  The noise floor was not improved by the new amps, but it wasn't worse either.  My main complaint with the Sontec amps was that they occasionally made low level hissing and crackling noises (even new ones!).  
I cut the test short when I discovered a harmonic of the power supply regulator on the right channel (very quiet, but there).  This was with the eq filters at max gain.  At first I thought the 990 was drawing too much current, but after I swapped the channel's op amps I realized that it was the power supply itself causing the problem.  On our Sontec the bridge rectifier, a regulator, and main filter caps are about 1-2 inches from the right channel's amps and bypass circuit.  The transformer is also on that side of the box.  I brought the eq back to the shop and powered it up with the bench supply.  No buzz.  So now I am going to power it remotely.  
Some other notes:
John Hardy said that his amp will work at 28 Volts but be a bit hot.  He wasn't joking.  It gets HOT.  It's probably a good idea for me to regulate the +-28V down to 24V in the end.

[James Craft]

A few years ago I was researching a 990 type opamp for a mic-pre and I can't remember who it was, but they had different "voltage" models of their opamp listed that were available anywhere from +/-15 through +/-30. I found this on the web, so I'll take a look and see what I can come up with, I thought it was Forssell, but I just checked his page and specs for the 992 only go to +/-24.

[Greg Norman]

I'm thinking of trying the OPA134AP for the input amp.  It's an all FET input amp that looks good on paper.  Not so impressed with the headroom limitations, but am still curious.  Does anyone have experience with it?
http://focus.ti.com/docs/prod/folders/print/opa134.html

[Viitalahde]

The 134 is a little grainy sounding in my opinion, though its specs might look better. They also take +/-18v at max only, if I remember it right.

I really suggest you to try the OPA604, or the OPA627 (which also operates at max. +/-18v), which is a sweet thing!

[Dan Kennedy]

Decent quality opamps (5534, 275, 604) exceed the performance of
most circuit board layouts by a rather large margin.

If you haven't actually put them in a rigoursly laid out test bench, you know nothing.

If you don't translate that knowledge to a reasonable facsimile of that circuit arrangement to the circuit board of choice, you still don't know nothing.

You can't make any judgements by "pick and play" without knowing
a lot more than "it passes audio", which is what a whole slew of people think is enough.

I learned most of what I know now by doing this opamp swapping shit back when the 5534 first came out. I also learned why it worked in some places the TL-071 didn't and vice versa. 1974 for me. George knew it earlier.

Discrete opamps exceed the performance of monolythic opamps, but they do it because the designer has picked which of the specifications matter the most for the application, rather than the usual, which one can I buy that will get me closest, cheapest.

Dan

[Greg Norman]

Quote:

You can't make any judgements by "pick and play" without knowing a lot more than "it passes audio", which is what a whole slew of people think is enough. {snip} Discrete opamps exceed the performance of monolythic opamps, but they do it because the designer has picked which of the specifications matter the most for the application, rather than the usual, which one can I buy that will get me closest, cheapest.

I can understand your sentiment, however compared to the standard, a 741 op amp from radio shack would do better in this eq.  The circuit requirements are not too foreign, and my demands are little.  As "musical" and great sounding the sontec amp may be, they are unusable in a real recording session.
So the standard is shit.  We need to find better than shit.  This eq has been out of the lineup for quite a while, and I could have sold it and bought a better product.  I am choosing to use what I have though.  There is plenty there, but not enough to be confident about.   I realize that in my daily work I'm not going to break any records, however I must get a good sounding, useful eq out of this box.  I'm not too concerned about having a slew rate that will lignite diesel fuel, or a bandwidth that will transmit to the stars.  I'm  rebuilding a faulty section of an otherwise great eq.  This is not a muscle car in the garage for me.  It can be made good within a reasonable try.

Quote:

If you haven't actually put them in a rigoursly laid out test bench, you know nothing.

I know that.  That is why I'm asking for opinions.  I don't have time to try out 16 different op amps on a "could be good if it worked" situation.  That is why we are here it seems.  If I knew exactly what I needed to do, I wouldn't be posting anything unless someone needed info.  If someone knew that the OPA134 amp was unstable, or plainly poor, they could tell me.  Like I have experienced problems with some amps in the past.
In the end, we hopefully end up with a good solution for wayward owners.  


By the way.  Your preamps are great.

[Dan Kennedy]

The problem is too many people just drop opamps into any old where, and many times I've had to go back and undo it, or spend a fair amount of time dealing with stability issues that cause audible distortions when knobs get set just so.

Or bipolars in a high-Z circuit so switches start clicking or pots crackling.

Mostly stability problems.

I personally like the OPA627 as a very smooth musical opamp. The headroom might not be that big a deal, or you could pad the signal a bit on the way in, and use the 6db of gain in a balanced output stage.

I do hope you have a scope to verify there isn't anything to gross going on...

[moogus]

George Massenburg wrote on Sat, 10 July 2004 10:16

You're on the right path, except that there're aren't many chips that can run +/- 28v.  Although the Hardy goes to +24v doesn't it? George

Our 99V opamp runs on +/-34V, if thats any help.


M