dual Vs single opamp for stereo mic preamp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

spaceludwig

Well-known member
Joined
Jul 14, 2011
Messages
186
Question for the pros...

I would like to build a stereo mic pre based on the Neve IC pre found on the JLM audio website. I modified the design slightly years ago (even had a thread on the old website) and would be happy to post the schematic once I complete the layout.

Since I need to amplify a stereo microphone I would need to build 2 channels. My question is the following: Other than cost, space, part count, and current draw are there any reasons to use 2 single opamps rather than just one dual opamp (i.e. two 5534 vs one 5532)? Is there a risk of additional noise entering either signal - or anything else that would degrade audio quality or functionality -  when using a dual opamp?

My apologies in advance if this is a silly question, I am not knowledgeable enough to figure this one out. I have breadboarded one channel and it sounds fine to my ears. Any info/advice/suggestions would be much appreciated.

Cheers!

EDIT: Attached is one channel of schematic. Amplification stage based on JLM's hybrid opamp. Could not find symbol for DRV134 balanced driver so made my own which is admittedly very crappy but I've never used schematic program so my first attempt.
 

Attachments

  • Stereo Preamp.png
    Stereo Preamp.png
    13.3 KB · Views: 111
The general answer is that dual opamps are generally equivalent to their single versions, with excellent separation and crosstalk specification between halves.

The specific example you offer, 5532 vs 5534 is one example that breaks that rule.

The 5532 is internally compensated to be unity gain stable, while the 5534 is "decompensated" or under compensated so only stable for gains above 10 dB or so, unless you provide adequate external compensation capacitance.

The decompensated 5534 provides slightly faster slew rate, and slightly more open loop gain, if your design lets you take advantage of that (nose gain > 3x).  Noise gain = the voltage gain WRT the + opamp input.

JR


 
Thanks for your answers. I just gave the 5532/34 as an example. I am thinking of using the opa2604 or perhaps the opa2134.

@abbey road d enfer

Could you please offer some general advice regarding layout as it pertains to this particular project? I'd be happy to post PCB layout for critique once I start.
 
Quick question regarding forum protocol:

Should I start a new thread every time I have specific questions pertaining to this particular project or should I start one overall thread?  I'm trying to balance getting as much qualified feedback (npi) as possible without annoying everyone with multiple threads or violating posting etiquette.
 
Just keep babbling in here, that way it's in one thread, then maybe rename the thread to "opamp pcb layout Q&A" or something along the general theme of the thread... Having a bunch of little threads makes for bits to get lost easily.
 
Paul, at the risk of opening a can of worms, would you mind listing a few you feel make that list?

Thanks in advance!
 
spaceludwig said:
Paul, at the risk of opening a can of worms, would you mind listing a few you feel make that list?

The LT1128 leads the list; also the OPA551 (which I wouldn't use for a preamp, but that's a different issue). Some other useful opamps not available as duals: LT1007, LT1115.  And while the LT1468 is theoretically available as the dual LT1469, if you want the through-hole version rather than the surface-mount you're S.O.L. unless you order thousands of them (or beg Linear Technology for samples -- maximum 2). In practice, I call that unobtainable. 

Add the OPA627 and OPA637 (their prices are also sky-high), and most of the chips from Analog Devices.

Peace,
Paul

 
Thanks all for the replies...

Since I have no practical experience (other than the the lab i have fashioned) I will stick with what I have.

My findings so far are that having a transformer to couple the input signal to the gain opamp is much, much quieter than using a INA103 as a replacement for an input transformer. The noise is overbearing. Perhaps I lack practical experience to use it properly. Please advise....
 
416ts7JvW0L._SL500_AA300_.jpg
 
spaceludwig said:
My findings so far are that having a transformer to couple the input signal to the gain opamp is much, much quieter than using a INA103 as a replacement for an input transformer.
INA103 isn't a replacement for a transformer.  It replaces BOTH the transformer and the following OPA.

You need to post circuits of both versions if you want us to understand what you  are trying to do.
 
Hi Ricardo,

I know it's an all-in-one package but I thought why not try it as a transformer substitute with a gain of like 4 or 5. I just found it wasn't very quiet, i.e. at full gain there was a noticeable shhhh sound unlike when a transformer is being used.

Some more info about the project:

I posted a schematic of the circuit in the first post.

I will be using some of the tamura input  transformers I pulled out of the PM-700 console I upgraded last summer. They are 600:3K and are relatively smallish. The 1uF/1.6K zobel at the input is what I got using the scope with a square signal. On the PM-700 input modules they use 1uf/47K but that value just attenuated way too much of the signal when I had it on the breadboard. It would also appear that using different opamps requires changing the resistor value in the network to maintain signal integrity.

Could someone explain what the the 100pf/510 RC network in the feedback loop is doing? Is this a filter of sorts? It would seem the cutoff is 3Mhz, though I'm probably way off in my iterpretation...

If anyone has any comments or suggestions with this design - which is basically the Neve IC pre from JLM audio using his Hybrid opamp design, please feel free to share.

 
OK, am now working on PCB which should be amusing, if anything, since I don't know the first thing about making them. Hopefully you all will be feeling charitable when I post my first efforts and you are overcome with the urge of hurling rotten tomatoes... ha
 
spaceludwig said:
Hi Ricardo,

I know it's an all-in-one package but I thought why not try it as a transformer substitute with a gain of like 4 or 5. I just found it wasn't very quiet, i.e. at full gain there was a noticeable shhhh sound unlike when a transformer is being used.

The reason for the noise is that single-chip preamps like this actually have a fairly high equivalent input noise when set for low gain. Under normal circumstances, when the preamp chip is the whole circuit, that's not a big deal; you'd only use it in the noisy low-gain setting when the mic signal was really large, and the signal-to-noise would still be decent.

However, if you try using it in low-gain configuration to replace a transformer, you're using it in its noisy mode for lower-level mic signals, and you'll hear the noise.

Some more info about the project:

I posted a schematic of the circuit in the first post.

I will be using some of the tamura input  transformers I pulled out of the PM-700 console I upgraded last summer. They are 600:3K and are relatively smallish. The 1uF/1.6K zobel at the input is what I got using the scope with a square signal. On the PM-700 input modules they use 1uf/47K but that value just attenuated way too much of the signal when I had it on the breadboard.

That's odd; it shouldn't do that. Try it with the 20k load resistor plus the original Zobel network, and watch out for stray capacitances. Make sure your signal generator's output impedance is 150 ohms. Also make sure your 20k and 47k resistors are really the right values; I had a 68k resistor measure about 80k in last Thursday's workshop session. (What? You mean sitting around in my basement for 30 years actually changes resistor values?)

It would also appear that using different opamps requires changing the resistor value in the network to maintain signal integrity.

That also shouldn't be happening. The opamp theoretically shouldn't be loading the transformer at all, so which opamp it is shouldn't affect the loading, and the network should remain the same.

Could someone explain what the the 100pf/510 RC network in the feedback loop is doing? Is this a filter of sorts? It would seem the cutoff is 3Mhz, though I'm probably way off in my iterpretation...

Yes, it's a filter; actually a step filter, which begins rolling off at 15.9kHz (C1-R5) and stops at about 3MHz, as you calculated (C1-R6). It's mostly to flatten the response of the transformer+opamp system, which probably has a bit of a peak up top even with the Yamaha Zobel network. That used to be common practice -- design a transformer with a bit of a peak, then flatten it out with a network in the amplifier circuit.

If anyone has any comments or suggestions with this design - which is basically the Neve IC pre from JLM audio using his Hybrid opamp design, please feel free to share.

If that transformer is rated at 600:3000 ohms, then with a 150 ohm microphone it'll have an output impedance of 750 ohms -- 722 ohms, if you count the 20k resistor in parallel with the secondary. So you want an opamp with an ideal source impedance close to that, and the obvious choice is an LT1128; with e(n) = 0.85nV and i(n) = 1pA, ideal input Z = e(n) / i(n), or 850 ohms. That's plenty close enough. The LT1128 is also very low distortion and has better-than-average DC performance for a bipolar opamp. Digi-Key has 'em.

Peace,
Paul
 
ricardo said:
pstamler said:
... loadsa good stuff ...
I second all this from Guru Paul.

My only caveat is that LT1028/1128 have evil input bias cancelling so the noise performance is usually slightly worse than you'd expect from the spec.  especially in transformer type circuits like yours.
 
Hi Paul, thanks for taking the time to answer so thoroughly. I appreciate the extra effort you`ve taken to explain things in some detail.

I do have some additional questions. If you find the time to answer I assure you I do not just glance through them.

Make sure your signal generator's output impedance is 150 ohms

It is 600 ohm. Could I buffer the signal with an opamp to lower the impedance then fine tune with a resistor in series with the output?

That also shouldn't be happening.

I did not have the 20k load. Also, it only happened when I turned the circuit on.

Yes, it's a filter

Out of curiosity: I have read that higher resistance have higher noise. Would there be any value in increasing the Capacitor to say .1uF and lowering the value of the resistor to say 100 Ohm and 5 ohm. admittedly the filter won`t be the same value but I`m just wondering if the total noise would be lowered.

Alternately, is there any harm in getting rid of the filter altogether?



 
My only caveat is that LT1028/1128 have evil input bias cancelling so the noise performance is usually slightly worse than you'd expect from the spec.  especially in transformer type circuits like yours.

Thanks for the heads up. Unfortunately I do not understand what input bias cancelling is. I have a rudimentary understanding of feedback amp but I`ll have to research this one. I`d be happy to read a layman`s explanation though.
 
spaceludwig said:
Hi Paul, thanks for taking the time to answer so thoroughly. I appreciate the extra effort you`ve taken to explain things in some detail.

I do have some additional questions. If you find the time to answer I assure you I do not just glance through them.

Make sure your signal generator's output impedance is 150 ohms

It is 600 ohm. Could I buffer the signal with an opamp to lower the impedance then fine tune with a resistor in series with the output?
Yes, that would work but it is even easier to add a passive pad (resistor divider) that scubs off signal level and delivers a useful source impedance.
That also shouldn't be happening.

I did not have the 20k load. Also, it only happened when I turned the circuit on.

Yes, it's a filter

Out of curiosity: I have read that higher resistance have higher noise. Would there be any value in increasing the Capacitor to say .1uF and lowering the value of the resistor to say 100 Ohm and 5 ohm. admittedly the filter won`t be the same value but I`m just wondering if the total noise would be lowered.

Alternately, is there any harm in getting rid of the filter altogether?

?? There is thermal noise in resistors that increases with value.

Any input filter is in parallel with source impedance (150-200 ohm) and input termination (1.5-2k). You don't want the input filter to step on the nominal 10x resistive input termination (1.5-2k ohm).

It is generally a good Idea to filter out RF before high voltage gain stages to prevent rectification and receiving radio stations in your preamp. If you have a quiet location and well shielded sources you might forgo an input filter, but I wouldn't.

JR
 
Back
Top