Mid Freq EQ - Possible replacements for LM741

Hi,
I would like to improve the IC that it's on the Mid Frequency EQ of an Alice 828 (first version with Belclere transformer and Discrete Mic Pre).

It has been discussed a lot on the possible replacements for the LM741 being the normal candidates:

TL071CP (low noise version of the TL071, pin compatible, low current draw)
NE5534AN by Signetics  (not pin compatible, might need compensation caps, more current draw)
Burr Brown OPA 134A (pin compatible with 741)

I know that depending on the circuit usage of the 741 sometime it's  indifferent to upgrade it.
Not understanding the circuit very well I would like to ask if in this particular circuit upgrading the LM741 would be justified?
And if you advise on the TL071CP or OPA 134A for the upgrade?

Thank you

here is the schematic:

Whoops said:

Not understanding the circuit very well I would like to ask if in this particular circuit upgrading the LM741 would be justified?
And if you advise on the TL071CP or OPA 134A for the upgrade?

Click to expand...

my suggestions, in order of preference:
TL071
LF356
LF351

The single BJT Baxandall in front of that begs for a re-think. The one part is doing more than all the 27 transistors inside the '741.

PRR said:

The single BJT Baxandall in front of that begs for a re-think. The one part is doing more than all the 27 transistors inside the '741.

Click to expand...

Hi PRR, could you please explain a bit more

What problem do you see?

thanks

Whoops said:

Hi,
I would like to improve the IC that it's on the Mid Frequency EQ of an Alice 828 (first version with Belclere transformer and Discrete Mic Pre).

It has been discussed a lot on the possible replacements for the LM741 being the normal candidates:

TL071CP (low noise version of the TL071, pin compatible, low current draw)
NE5534AN by Signetics  (not pin compatible, might need compensation caps, more current draw)
Burr Brown OPA 134A (pin compatible with 741)

I know that depending on the circuit usage of the 741 sometime it's  indifferent to upgrade it.
Not understanding the circuit very well I would like to ask if in this particular circuit upgrading the LM741 would be justified?
And if you advise on the TL071CP or OPA 134A for the upgrade?

Click to expand...

I would recommend OPA134, because it is very tolerant on its environment (layout, rail decoupling) and doesn't draw more current, whilst being faster, less noisy and have better output capability, although these two performance improvements won't really show, because of the limitations of the surrounding circuitry. Being used in what is esentially a gain of 2 configuration, noise is not an issue, and regarding output capability, the preceding circuitry is not very capable; the topology of the mic preamp is a serious headroom-limiting factor. Any attempt to significantly improve performance implies a major redesign.

abbey road d enfer said:

I would recommend OPA134, because it is very tolerant on its environment (layout, rail decoupling) and doesn't draw more current, whilst being faster, less noisy and have better output capability, although these two performance improvements won't really show, because of the limitations of the surrounding circuitry. Being used in what is esentially a gain of 2 configuration, noise is not an issue, and regarding output capability, the preceding circuitry is not very capable; the topology of the mic preamp is a serious headroom-limiting factor. Any attempt to significantly improve performance implies a major redesign.

Click to expand...

Thanks Abbey and PRR

This is a compact mixer from the late 70's, Im sure it was on the inexpensive range at it's time.
It was designed by Ted Fletcher

Theres 2 versions of the console circuit wise,  although both have the same outside looks.
Theres the MKI with Belclere input Tx and discrete mic pre and Master section
and the MKII that can have Belclere or sowter input tx, 5534 based Mic pre and has an IC based Master section

There's no service notes for the MKI version, the one I have, so the schematic I show you was drawn by hand by Ted Fletcher lately and he was glad enough to send it to me.
For the MKII there's complete factory documents available.

I still didn't completely check the schematic against the circuit and I dont know if Ted did that or not,
so basically the schematic can have errors.
I will check it against the circuit and possibly re-draw it in Eagle.

Even if there's no errors in the schematic, I understand that from the engineering point of view the circuit is crude and has limitation, but I have to be honest I bought the console because I really like the sound.
I was producing a record  for a Rock band and an engineer friend lend me it's Alice 828, I used it on that record and I loved the sound for rock music. I also distorted the main vocals on some of the channels the way I usually do it on an SSL 4K and loved the results with the 828.
I like the crunch and the grain  it has, I also like the crude Limiter on the master section.
After that record I had to get one for myself, so I bought one, and I use it as one more tool when I'm doing rock music records.

After having the console I searched on the web and found that theres a lot of people that also likes this little console.

I guess I have to re-draw and re-check the schematic to be sure I'm not misleading you with a wrong circuit.

This is some info on the circuit and console from Ted Fletcher forums,
you are welcome to comment this,
thank you.


The 828 was our most successful design... we sold hundreds of them with customers ranging from the Singapore Army to Wimbledon Tennis Club. It was the classic of its day, a simple 8-channel transformer input mixer with a very useful stereo limiter compressor on its output.

There are many 'gee-wizz' stories about 828s; one of my favourites actually happened to me.... on a visit to Pinewood I was shown the brand new (as it was then) sound stage with its outrageous Neve mixer that required at least 3 operators, and in the centre on the script area was an 828...... they were doing the final mix for the Bond movie 'Thunderball' through it because they liked the limiters!

CIRCUIT
Technically, the 828 was very interesting in hindsight.... the channel retained a transformer for the mic in; it was made for us by Dr Sowter. The channel amplifier was extremely simple with minimal feedback and running in class-A, and the mixing, very unusually, was passive! The result was a signal path with virtually no odd-order distortion (class-A and minimal feedback), the noise at the output was not great by modern standards, but was certainly good enough for professional studio use.
With its very high input overload margin, the performance was perfect for speech applications and it became the preferred mixer for radio and television O/B use as well as being ideal for commercial production.

The 828 was living proof of a number of 'ground rules' about quality sound reproduction.....
Overload margin (headroom) is everything.
Noise is relative.
The unpleasantness of distortion has little to do with the numbers.
And now after all these years, it was proof that low impulsive distortion sounds right!


the 828 was really simple....  transformer mic input, simple descrete mic amp, a Baxendall EQ with an added coil and mid section, and a really basic transistor limiter (similar to the AGC on a Philips cassette machine). The EQ was a single transistor(!!) The mixing was resistive; which suited the limiter, which needed a very low level.
Bear in mind, this was 1972 so things were a little primitive.... but it did sound good!

The original Alice 828 mixer was derived from an earlier type.... the Alice AD62.
In the early part of 1977 we decided to market a rugged but portable 8-channel mixer based on experience with the earlier lightweight type.
The circuitry was very professional for its price.... the mic amp was particularly good, using a Sowter or Belclere transformer into a very sophisticated little amplifier. ( I now have a sketch of the circuit of the Mk 1 channel board available).
We started selling the Mk 1 828 in July 1977 but rapidly altered the design of the mid EQ circuit as the very early few were unreliable; I incorporated an IC (!) amplifier stage and a discrete transistor tuned circuit which all worked very well.
We sold the Mk1 828 until mid 1982 when we introduced the Mk2 with completely re-designed circuitry making use of modern IC types. (Circuits available).

Under test, the distortion figures of the Mk 2 are better than the Mk 1, but actually, the sound of the Mk 1 is certainly sweeter..... but beware here.... the Mk 2 is still significantly superior to any present day 'economy' mixer.

The main secret with all these things is overload margin; these old desks were built to specifications demanded by broadcast people for use in 'outside broadcasts' which could mean a rapid set-up in a muddy field with no run-through!
Modern desks are designed so that the noise level is very low; inevitably this means that there is poor overload margin, so the thing sounds thin and nasty. These old 828s sound wonderful... and that's why there are so many still in use by discerning engineers in spite of their extreme age!

I certainly would not deny the sonic benefits of class-A operation, but the claims made about headroom must be taken with a pinch of salt.
With a single 22V rail and a two-transistor topology, the only way to get headroom is by operating the mixer at lower level than currently usual. Most mixers operate at a nominal operating level of ca. -2 to 0 dBu, which results in about 22 dB of headroom with bipolar 15 to 18V rails. The published schemo would need to operate at about -12 dBu for the same headroom. Combined with the proudly exposed passive mixing, this puts restrictions on the achievable overall noise performance. OTOH, the small number of channels (and possible clever use of discrete transistors in the summing amps) is an improving factor. Then I'm not surprized Mr; Fletcher admits a noise issue.
And he may have found serendipity in his admittedly simplistic limiter.
All is fair and well, except when Mr. Fletcher profers "Modern desks are designed so that the noise level is very low; inevitably this means that there is poor overload margin, so the thing sounds thin and nasty." Any analysis of a "modern" (later than 1975 +/- whatever) decently designed studio or broadcast mixer would deny this assertion.
It seems that his commercial activities lead him to embellish his past, which is part of the game, but when he attacks his possible competitors, he crosses the barriers of ethics.

abbey road d enfer said:

I certainly would not deny the sonic benefits of class-A operation, but the claims made about headroom must be taken with a pinch of salt.
With a single 22V rail and a two-transistor topology, the only way to get headroom is by operating the mixer at lower level than currently usual. Most mixers operate at a nominal operating level of ca. -2 to 0 dBu, which results in about 22 dB of headroom with bipolar 15 to 18V rails. The published schemo would need to operate at about -12 dBu for the same headroom. Combined with the proudly exposed passive mixing, this puts restrictions on the achievable overall noise performance. OTOH, the small number of channels (and possible clever use of discrete transistors in the summing amps) is an improving factor. Then I'm not surprized Mr; Fletcher admits a noise issue.
And he may have found serendipity in his admittedly simplistic limiter.
All is fair and well, except when Mr. Fletcher profers "Modern desks are designed so that the noise level is very low; inevitably this means that there is poor overload margin, so the thing sounds thin and nasty." Any analysis of a "modern" (later than 1975 +/- whatever) decently designed studio or broadcast mixer would deny this assertion.
It seems that his commercial activities lead him to embellish his past, which is part of the game, but when he attacks his possible competitors, he crosses the barriers of ethics.

Click to expand...

It seems to be the nature of the mixer business to bad mouth everybody else, and remember the past fondly.

I was impressed by a few of Ted Fletchers later circuit blocks I've seen around the WWW, that channel strip not so much.

If you are pursuing a legacy (old) sound why replace the 741? With only 22V rail if will take 44uSec to slew rail to rail... This is perhaps an 8-10kHz power bandwidth (check my math), so shouldn't slew rate limit with normal musical signals (HF content is usually -10dB from LF). The 741 may not even be the weakest link in that chain.

I think Steve Dove may have done some (one?) mixer work with Alice in the '80s (IIRC a broadcast mixer but don't know any details, Steve just mentioned it once in passing at a trade show). 

JR

abbey road d enfer said:

All is fair and well, except when Mr. Fletcher profers "Modern desks are designed so that the noise level is very low; inevitably this means that there is poor overload margin, so the thing sounds thin and nasty."

Click to expand...

Abbey I understand what you mean, but in that comment  Mr Fletcher was talking about Modern Low end economic compact mixers, the target of the alice 828 at it's time, and not professional high end mixer.
We are talking more in the Mackie, Behringer, Yamaha MG type of mixers, that actually sound thin and nasty (Mackie being much better than the other but cheap none the less), quite the opposite of the Alice 828 sound.

Thank you

Whoops said:

Abbey I understand what you mean, but in that commend I Mr Fletcher was talking about Modern Low end economic compact mixers, the target of the alice 828 at it's time, and not professional high end mixer.
We are talking more in the Mackie, Behringer, Yamaha MG type of mixers, that actually sound thin and nasty (Mackie being much better than the other but cheap none the less), quite the opposite of the Alice 828 sound.

Thank you

Click to expand...

Please allow me to disagree; I've used a number of those cheap mixers and most of the time the performance was adequate; inversely I've heard bas sounds from reputed boards, operated by morons. You would be surprised at the little difference there is between a cheap mixer and an expensive one in terms of the basic circuitry. The circuitry in the Alice mixer is not very different than that of a contemporary Helios; both share the same goods and bads.

JohnRoberts said:

It seems to be the nature of the mixer business to bad mouth everybody else, and remember the past fondly.

Click to expand...

Please JR read my last comment, I think Mr Fletcher was not bad mouthing anyone

JohnRoberts said:

If you are pursuing a legacy (old) sound why replace the 741? With only 22V rail if will take 44uSec to slew rail to rail... This is perhaps an 8-10kHz power bandwidth (check my math), so shouldn't slew rate limit with normal musical signals (HF content is usually -10dB from LF). The 741 may not even be the weakest link in that chain.

Click to expand...

Well John dont get me wrong, I'm not pursuing any legacy (old) sound, I really like the sound  of this console channels the way it is.
I don't know if it's old or new, it's just a nice tool.
The reasons I would like to replace the LM741 are:
- Because it's cheap and easy to stick an opamp with better spces
- Because changing it will not alter what I like about the Alice 828 channels sound
- Because I really dislike the LM741 and psychologically I will fell better if I spend 10 euros putting there something better, even if the benefits are marginal because the costs are also marginal.

I will not and have no intention in changing or modifying the channels circuit any more than that,
Maybe I will modify the Masters Output for a balanced output, but thats it.
I like how it sounds and works for my needs.

I totally understand that sometimes people think wrongly that they can improve a specific circuit performance by altering the Opamps, when thats not the case because of the surrounding circuit or topology.
So just let me tell you that I know that changing the Opamp in the EQ will not improve the performance, thats fine.

I was just asking if I could do it in this particular 741 implementation, so that i dont get into instability or oscilation problems,  and if you advised on the TL071 or OPA134.

Abbey advised on the OPA134 and gridcurrent on the TL071.

Thank you

abbey road d enfer said:

Please allow me to disagree; I've used a number of those cheap mixers and most of the time the performance was adequate; inversely I've heard bas sounds from reputed boards, operated by morons. You would be surprised at the little difference there is between a cheap mixer and an expensive one in terms of the basic circuitry. The circuitry in the Alice mixer is not very different than that of a contemporary Helios; both share the same goods and bads.

Click to expand...

Well I never said that performance of cheap mixers was not adequate.
I said that Behringer cheap compact mixers and Yamaha MG series mixers sound Thin and Nasty, thats my opinion.
Performance is adequate for the most people that use them.
And I said also that Thin and Nasty is not the sound of the Alice 828

Abbey my electronics knowledge is far from yours, thats why I'm always learning with your posts and tips.
This is an hobby to me and I love to learn with you guys, and I'm also able to help people that know less than me.
But I'm a very experienced Sound Engineer, I've used most of the Consoles around, both in Studio and Live sound, thats my everyday Job.I used both ends of the range from Behringer 60 euros mixer, to a Neve 8048, SSL 9000 , EMI TG, SSL4k.

For a cheap mixer the Alice doenst sound thin, and sounds quite interesting for my Rock duties and a Yamaha MG just sounds thin and nasty and I have no use for it.
So if Mr Fletcher was comparing he's own designed cheap mixer sound to any other modern cheap mixer, I think he's right the Alice 828 sounds much better, but with much higher noise also.






 

abbey road d enfer said:

Please allow me to disagree; I've used a number of those cheap mixers and most of the time the performance was adequate; inversely I've heard bas sounds from reputed boards, operated by morons. You would be surprised at the little difference there is between a cheap mixer and an expensive one in terms of the basic circuitry. The circuitry in the Alice mixer is not very different than that of a contemporary Helios; both share the same goods and bads.

Click to expand...

+1  by the 1980s modern ICs were so much faster, quieter, and lower distortion than audio needed that inexpensive audio paths were pretty much on par with premium audio paths.  I spent decades speculating about what the real differences were between console brands that attracted user loyalty. For Mackie it was millions of dollars spent on advertising.  :

Subtle things like gain law, EQ centers, Q, and boost/cut control response affect the perceived sound.  This is not objective science  so to some degree art, and the most successful console makers have developed an ergonomic  recipe that works for them and their customers. 

Legend and folklore is part of that brand halo.

JR

PS: It might be interesting to make some baseline bench measurements before and after swapping out the 741 for say an OP134(?). 

JohnRoberts said:

+1  by the 1980s modern ICs were so much faster, quieter, and lower distortion than audio needed that inexpensive audio paths were pretty much on par with premium audio paths. 

Click to expand...

Well at the moment her in the workshop I have channels of an SSL 4K E-Series , Neve 1073 , a Mackie VLZ 1604 mixer, and Alice 828, and a Behringer UB1002.

And I can tell you that soundwise the Behringer UB1002 and the Mackie 1604 are not on par with the SSL 4K channel or the Neve 1073 or the Alice 828.
The Alice is also not on par with the Neve 1073 and SSL 4K for sure, but it does it's own specific thing.

I'm assuming you consider Behringer UB1002 and the Mackie 1604 inexpensive audio paths and the Neve 1073 and SSL 4K premium audio paths.

Or probably we are talking about different things, but I really thing you got what Mr Fletcher said wrong.
And Abbey also misunderstood what I said.

Anyway thank you.
I will try to redraw the schematic.

Whoops said:

Well at the moment her in the workshop I have channels of an SSL 4K E-Series , Neve 1073 , a Mackie VLZ 1604 mixer, and Alice 828, and a Behringer UB1002.

Click to expand...

So you could compare frequency response, linearity (THD), S/N, etc. 

That might be revealing of what is different and how (much). 

JR

And I can tell you that soundwise the Behringer UB1002 and the Mackie 1604 are not on par with the SSL 4K channel or the Neve 1073 or the Alice 828.
The Alice is also not on par with the Neve 1073 and SSL 4K for sure, but it does it's own specific thing.

I'm assuming you consider Behringer UB1002 and the Mackie 1604 inexpensive audio paths and the Neve 1073 and SSL 4K premium audio paths.

Or probably we are talking about different things, but I really thing you got what Mr Fletcher said wrong.
And Abbey also misunderstood what I said.

Anyway thank you.
I will try to redraw the schematic.

Click to expand...

JohnRoberts said:

So you could compare frequency response, linearity (THD), S/N, etc. 

That might be revealing of what is different and how (much). 

JR

Click to expand...

You could do that if you want and show the results that would be nice.

I'm moving on to the topic. Thanks

JohnRoberts said:

If you are pursuing a legacy (old) sound why replace the 741?

Click to expand...

JR it's always really strange when people reply in a thread saying something that you never said or think, so your post was really confusing to me.

Now I got it, you clearly confused my thread with jdurango thread "sound of LM741 amp" that is in the Lab section also.
Jdurango a new member, has a console with 741, and has he  is persuing 50/60s music he was asking if he could achieve that sound with the 741 opamp.

https://groupdiy.com/index.php?topic=64551.0

Now it makes sense the out of context comment.

It happens

Whoops said:

Please JR read my last comment, I think Mr Fletcher was not bad mouthing anyone

Click to expand...

Well, he was indirectly, by relating low-noise design with "thin and nasty". In fact, I really would like to have an objective analysis of thin and nasty; thin may be an LF response issue, and nasty some kind of distortion. In that case that should be measurable. To my knowledge, even the cheapest mixers have quite extended LF response, so if the Alice has more bass, it would not be flat. I believe I could easily modify a mixer to boost bass and it would make any other mixer thin in comparison.
As to distortion, it is true that the topology used by Mr. Fletcher tends to produce a significant level of 2nd harmonic, which is known to be euphonic. It may make a mixer with double-zero distortion seem nasty in comparison.

- Because changing it will not alter what I like about the Alice 828 channels sound

Click to expand...

Who knows? Maybe what you like in the Alice's sound is the 741 sound?

- Because I really dislike the LM741 and psychologically

Click to expand...

I cannot really comment on this since it seems to me a matter of personal belief. Do you have any measurable evidence to sustain this opinion?

abbey road d enfer said:

Who knows? Maybe what you like in the Alice's sound is the 741 sound? 

Click to expand...

Well, if I'm saying that it's because I know, when I don't know I ask.
The EQ section was turned off

abbey road d enfer said:

I cannot really comment on this since it seems to me a matter of personal belief. Do you have any measurable evidence to sustain this opinion?

Click to expand...

You're right, just personal belief. I was completely honest about that.
I told my honest reasons.

Thanks Abbey