Replacing console fader amp with 2520 - circuit changes necessary?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

living sounds

Well-known member
Joined
Jul 26, 2006
Messages
3,931
Location
Cologne, Germany
Attached is the channel output section of my console. Everything is handled by IC5, a TL071 originally. I've replaced this with a NE5534, with added compensation cap and a 75 ohm series resistor after the op amp for stability.  Works fine.

I've got lot's of those nice 2520 DIY op amps with DIP8 connectors from our group member Abe (acsound) and I've tried  one of them in there. It seems to work more or less fine, though there is a 1.7 db drop in output amplitude and increased low frequency distortion.  I've dropped R46 to several lower values which increased the gain (as it should), but didn't otherwise change anything.  It seems to be stable, and offsets are within the confines of the datasheet, I matched the op amps input transistor pair closely.

Sonically there is a little more low end with the 2520 (measureably),  and the typical dynamic distortion happens. It sounds pretty good, a little smoother on top and punchier in the midrange than the 5534.

So, is this all, or does the circuit require adjusting for the op amp to work correctly? Thanks!
 

Attachments

  • SC200fader.png
    SC200fader.png
    163.6 KB · Views: 233
I would be nervous about any op amp that deviates significantly from the closed loop gain as defined by the feedback network.  An op amps open loop gain can measurably influence closed loop gain if there isn't much loop gain margin (gain headroom).

While it is difficult to directly measure open loop gain on the bench one test that I used back in the '70s before op amp reliability increased to today's high standards was to operate an op amp as simple unity gain inverter (equal say 100k input and feedback resistors). Then add a 100 ohm or smaller resistor from the - input to ground. The op amp will think it is working at 60dB or more of gain, while actually delivering a unity gain output. Compare the frequency response, distortion, and phase shift with and without the 100 ohm resistor in place.  I used this test to screen out weak sisters and would cull out a small handful of  weak performers  from a typical 1000 piece order.

Of course if this non-ideal response is desirable to you, enjoy. At least this will help you characterize and understand how it is deviating.  Note: if the DC performance of the op amp is too bad to run 60dB of DC gain you can cap couple the 100 ohm resistor. You mention LF response errors so use a decent sized cap. 

JR
 
> there is a 1.7 db drop in output amplitude and increased low frequency distortion.

Any audio op-amp has gain at 1KHz over 5,000.

The stage here is rigged for gain of 3.55.

We expect gain-error to be <0.01dB.

Not 1.7dB.

An error of 1.7dB suggests an open-loop gain of 16, which is very unlikely for '07x, 5532/5534, or 2520, for any frequency below radio. It might be 16 at 500KHz, but 160 (0.2dB error) at 50KHz, and under 0.1dB across the audio band.

We also expect open-loop gain to rise, and closed-loop error and THD to fall, at lower frequencies. So increased LF distortion isn't reasonable.

5534 should not need "added compensation" to work at gain of 3.55.

The added 75 Ohms doesn't cause 1.7dB loss into the ~~1.4K loading, more like 0.5dB.

The higher bias currents of the bipolar inputs should not be a gain-drop in this circuit.

The 100pFd across 12K takes effect above 130KHz.

So something is Very Wrong.
 
living sounds said:
Attached is the channel output section of my console. Everything is handled by IC5, a TL071 originally. I've replaced this with a NE5534, with added compensation cap and a 75 ohm series resistor after the op amp for stability.  Works fine.
Can you show us EXACTLY how the 75R is connected?
 
Thank you both.

I've built a second op amp, and this time the gain stays the same. I'll have to check what's wrong with the first one.  Otherwise this second op amp behaves like the first one, higher 1st harmonics distortion than the IC op amp, but subjectively better sound.  Makes much more of a difference than I had expected in this position - it doesn't provide a lot of gain and/or drive a transformer directly after all.

What I meant by "increased low frequency distortion" was "increased vs. the 5534".

From what I've gathered the 5534 is only stable at gains of 5 and higher, so I added a 10pf cap between pins 5 and 8.

The added 75 ohms are there regardless of the new op amp, I had added them in before for the benefit of the 5534 for stability (Soundcraft also did this in other consoles where they used a 5534 instead of a TL071) as fader amp.

So as far as the 2520 is concerned, is there anything I should change/add regarding the console's circuit? Thanks!

 
abbey road d enfer said:
Can you show us EXACTLY how the 75R is connected?

After the feedback loop before C29. I realize Soundcraft put it inside the feedback loop in series with the feedback resistor,  but from what I've read what I did should increase stability, and was far easier to do. Would it be better to remove the resistor with the 2520 in mind?
 
living sounds said:
From what I've gathered the 5534 is only stable at gains of 5 and higher, so I added a 10pf cap between pins 5 and 8.

The 5534 is stable at closed loop gains of 10 dB  (3x) or more.  Of course this is nominal stability so extra  input pin, or output load  capacitance to ground could make a difference.

JR 
 
living sounds said:
abbey road d enfer said:
Can you show us EXACTLY how the 75R is connected?

After the feedback loop before C29. I realize Soundcraft put it inside the feedback loop in series with the feedback resistor,  but from what I've read what I did should increase stability, and was far easier to do. Would it be better to remove the resistor with the 2520 in mind?
I agree with PRR: "something is Very Wrong. ". I can't help thinking that the 75R is not connected as it should, like it is shorted to something.
 
JohnRoberts said:
The 5534 is stable at closed loop gains of 10 dB  (3x) or more.  Of course this is nominal stability so extra  input pin, or output load  capacitance to ground could make a difference.

JR

You guys were right, of course. Looks like I got my "information" from a thread at diyaudio.com...  ::)


So should something be changed to accomodate this op amp? The old API circuit specifications have a series cap between rg and ground. Would - say - a 100 uf cap inserted between R46 and ground lower THD here?
 
living sounds said:
So should something be changed to accomodate this op amp? The old API circuit specifications have a series cap between rg and ground.
  That is done to minimize DC offset due to input bias current. In the API, that is necessitated by the fact that the output drives a transformer, where DC creates distortion.
  Would - say - a 100 uf cap inserted between R46 and ground lower THD here?
If there is NO transformer connected at the output, it shouldn't make any difference.
 
living sounds said:
So as far as the 2520 is concerned, is there anything I should change/add regarding the console's circuit? Thanks!

There should be a cap (around 10uf  bipolar) between the fader and the + input of the op-amp to keep dc bias off of the fader wiper.

If the 75ohm resistor was within the feedback loop of the op-amp it would lower the output impedance and possibly help LR crosstalk through the pan circuit.

A cap between R46 and ground would keep the DC gain at unity.  Depending on the DC offset you were getting,  I would think it's best to leave it out.

It's not clear if the 10pf compensation cap between pins 5 and 8 is still in there.  If it is,  I would take it out.  It's not needed with the 5534 and maybe it's affecting the 2520.  Also,  could the dip 2520 be causing any stability issues?

C29 seems a little undersized.  I would check the polarity of the dc offset with each op-amp and maybe raise the value a bit.
 
Thanks abbey!

LRRec said:
There should be a cap (around 10uf  bipolar) between the fader and the + input of the op-amp to keep dc bias off of the fader wiper.

Apart from a slightly scratchy fader (doesn't bother me) are there any other adverse effects from leaving it out?

If the 75ohm resistor was within the feedback loop of the op-amp it would lower the output impedance and possibly help LR crosstalk through the pan circuit.

So would you change this? Does the 75 ohm resistor serve a purpose as it is with the 2520?

A cap between R46 and ground would keep the DC gain at unity.  Depending on the DC offset you were getting,  I would think it's best to leave it out.
Yes, DC offset is not a problem.

It's not clear if the 10pf compensation cap between pins 5 and 8 is still in there.  If it is,  I would take it out.  It's not needed with the 5534 and maybe it's affecting the 2520.  Also,  could the dip 2520 be causing any stability issues?
It's there, I'll take it out since it is unnecessary for any op amp. But it wouldn't affect the 2520, since the pin in question is not connected to anything. The 2520 appears to be very stable at any gain I've tried in this configuration. Mechanical stability is not ideal however.

C29 seems a little undersized.  I would check the polarity of the dc offset with each op-amp and maybe raise the value a bit.
I have modded C29 on all channels to 220uf caps connected positive back-to-back with a resistor-fed bias from the + lines to the junction and a drain resistor to ground connected there. This is how it was done in the SSL4000, reverse bias can be avoided this way, sounds better this way to my ears, too.

Thanks!
 
A scratchy fader should bother you, not only regarding sonics, but for the long term health of the fader. Leaving the cap out would change the DC source resistance seen at the op-amp as the fader is moved up and down. I'm not sure how much effect that would have on the circuit. Probably not much. Who knows?

The 75ohm resistor is there to decouple what the op-amp sees at it's output. There is probably a good reason Soundcraft didn't have it there in the first place.

It occurred to me that you are replacing a tlo71, which draws 1.4ma of dc current with a 2520 which draws 15ma. I don't know how many channels you are doing this to, but this will probably over tax the power supply if you have not addressed that already.

Why not just replace the tlo with a better fet op-amp, like an opa132 or something similar and call it a day and use the 2520's for the master section?



living sounds said:
Thanks abbey!

LRRec said:
There should be a cap (around 10uf  bipolar) between the fader and the + input of the op-amp to keep dc bias off of the fader wiper.

Apart from a slightly scratchy fader (doesn't bother me) are there any other adverse effects from leaving it out?

If the 75ohm resistor was within the feedback loop of the op-amp it would lower the output impedance and possibly help LR crosstalk through the pan circuit.

So would you change this? Does the 75 ohm resistor serve a purpose as it is with the 2520?

A cap between R46 and ground would keep the DC gain at unity.  Depending on the DC offset you were getting,  I would think it's best to leave it out.
Yes, DC offset is not a problem.

It's not clear if the 10pf compensation cap between pins 5 and 8 is still in there.  If it is,  I would take it out.  It's not needed with the 5534 and maybe it's affecting the 2520.  Also,  could the dip 2520 be causing any stability issues?
It's there, I'll take it out since it is unnecessary for any op amp. But it wouldn't affect the 2520, since the pin in question is not connected to anything. The 2520 appears to be very stable at any gain I've tried in this configuration. Mechanical stability is not ideal however.

C29 seems a little undersized.  I would check the polarity of the dc offset with each op-amp and maybe raise the value a bit.
I have modded C29 on all channels to 220uf caps connected positive back-to-back with a resistor-fed bias from the + lines to the junction and a drain resistor to ground connected there. This is how it was done in the SSL4000, reverse bias can be avoided this way, sounds better this way to my ears, too.

Thanks!
 
+1 you don't want to allow DC to make a pot sound scratchy. I recall hearing  (somewhere) that over time this DC could damage the pot. I've never seen one damaged this way, but I've never left DC on one long enough to find out. 

JR
 
Thanks!

Soundcraft never put a resistor between fader wiper and op amp input since input DC offset is not a problem with a FET op amp like the TL071, right?

I'll take the 75 ohm resistor out then.

The PSU can handle the current, it's not the original one.

I've tried lot's of different op amps, and did not like how they sounded there. The 5534 was the best, and it measured well. But the 2520 sounds much more interesting, and I've got APP992 in the master section, sounds very nice this way.

LRRec said:
A scratchy fader should bother you, not only regarding sonics, but for the long term health of the fader. Leaving the cap out would change the DC source resistance seen at the op-amp as the fader is moved up and down. I'm not sure how much effect that would have on the circuit. Probably not much. Who knows?

The 75ohm resistor is there to decouple what the op-amp sees at it's output. There is probably a good reason Soundcraft didn't have it there in the first place.

It occurred to me that you are replacing a tlo71, which draws 1.4ma of dc current with a 2520 which draws 15ma. I don't know how many channels you are doing this to, but this will probably over tax the power supply if you have not addressed that already.

Why not just replace the tlo with a better fet op-amp, like an opa132 or something similar and call it a day and use the 2520's for the master section?
 
living sounds said:
Thanks!

Soundcraft never put a resistor between fader wiper and op amp input since input DC offset is not a problem with a FET op amp like the TL071, right?
Input bias current is a non-issue for BiFET op amps. There is reportedly a subtle distortion mechanism associated with imbalanced DC resistance at + and - inputs.  The typical post fader gain stage will have changing + input resistance with fader position so perhaps a resistor in series with the wiper can reduce the extreme resistance differences.... or not this is a little esoteric so feel free to ignore.

JR
 
living sounds said:
Soundcraft never put a resistor between fader wiper and op amp input since input DC offset is not a problem with a FET op amp like the TL071, right?
When Soundcraft designed the 500/600, they decided TL0's weren't good enough for post-fader amp, so they used 5534 in that position. They obviously were aware of the risks of noise due to bias current, so they put a capacitor between the wiper and the + input and a resistor to ground, but customers did complain because the faders were more noisy than the previous designs. They thought the faders were of poorer quality; some replaced them with P&G's, to no avail.  The cause was noise modulation. In BiFET's input noise current is almost inexistant but with bipolars it's significant. Varying the source impedance (which is what happens when you move the fader) changes the noise level. Although this variation is almost unmeasurable, it is clearly audible.
 
abbey road d enfer said:
JohnRoberts said:
Input bias current is a non-issue for BiFET op amps. There is reportedly a subtle distortion mechanism associated with imbalanced DC resistance at + and - inputs.
With BiFET's? Never heard about this... Do you have a reference?

http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=703

http://www.ti.com/lit/an/slyt595/slyt595.pdf

Sorry for the veer... and I wouldn't lose too much sleep over this for your post fader gain stage.

There is a subtle affect related to input capacitance that changes with input voltage. Apparently the old BiFets (like TL07x) were susceptible to this mechanism.

JR
 
abbey road d enfer said:
living sounds said:
Soundcraft never put a resistor between fader wiper and op amp input since input DC offset is not a problem with a FET op amp like the TL071, right?
When Soundcraft designed the 500/600, they decided TL0's weren't good enough for post-fader amp, so they used 5534 in that position. They obviously were aware of the risks of noise due to bias current, so they put a capacitor between the wiper and the + input and a resistor to ground, but customers did complain because the faders were more noisy than the previous designs. They thought the faders were of poorer quality; some replaced them with P&G's, to no avail.  The cause was noise modulation. In BiFET's input noise current is almost inexistant but with bipolars it's significant. Varying the source impedance (which is what happens when you move the fader) changes the noise level. Although this variation is almost unmeasurable, it is clearly audible.

The .4-1.5 pA of noise current in a 5534 is all of 1-3 nV/rt Hz of noise contribution worst case (2.5k from 10k fader) .  This is in addition to the roughly 3k impedance at the - input.

This might (?) be audible in a WFO listening test (I try not to argue about what other people hear),  but I would expect other noise sources in that complete audio path to dominate.

JR
 
Back
Top