Idea of an simple mic pre.

[thor.zmt]

I'll put it into TINA tomorrow when I need a break from other.

Ok, I put it in (schematics attached). I used 2SK209 models for the J-Fet's and BC550/560C models for the BJT's.

The J-Fet's have a bit higher Gm than the originals, the BJT's have lower Beta than the originals so I think the differences will mostly wash out.

Result for open loop gain:


The original circuit offers 94dB open loop gain with appx. 1kHz -3dB point (including Mr Miller). Uncompensated we have around 10kHz -3dB. going to 600 Ohm loading drops open loop gain a few dB, nothing major.

Closed loop gain for the M-15B (which I used as it has more details in the service manual) ranges from ~ 6dB to 56dB, so NFB ranges from 38dB minimum to 88dB at minimum gain.

My suggestion of Mosfet VAS with bootstrap load matches the original above 3kHz and increases LF gain (@ 100Hz to 114dB or 20dB more than original).

With a Darlington VAS + additional input stage degeneration (100R per J-Fet) overall gain is lower but LF gain is boosted.

With gain set to 12.7dB (with input transformer this gives ~ 33dB Gain) the closed loop frequency response is identical. Unless stated this setting (gain control @ 50%)

With 300mV output (nominal signal level according to Manual) and 10kHz all three circuits simulate as ~0.0003% THD with H2 being the only observable harmonic above the "noisefloor" of the sim and the original circuit being marginally superior to both modifications.

Raising signal by 20dB (still 10kHz) gives the original circuit 0.0017% THD with H2 & H3 notable.

The Mosfet version (e.g. mine) has 0.0022% THD and the lowest H3 (~3dB less than Darlington/degenerated and > 6dB lower than original), with Darlington inbetween. Going to 600 Ohm loading does degrade THD somewhat, around 3dB or so.

Simulations at 1kHz or 100Hz would likely favour the modified designs, but the value of these advantages may be questionable.

For noise, the darlington VAS is marginally worse than the other two, nothing I'd loose sleep over. The observed 4nV|/Hz @ 1kHz Ein are not that great, however we need to remember that the 1:10 step-up transformer on the input would lower this tenfold.

Unweighted SNR at maximum gain and 0.3V output level looks -60dB. Note that this is at 56dB Gain (actually 76dB total gain including transformer), so that is a helluvalot gain!.

Realistically, the original circuits offers fine performance that gives up very little to modern monolithic Op-Amp's in the way it is used, despite the very simple and unfashionable design. Non of the suggested "improvements", mine included appear to offer any material improvement.

As is, in the original context it is a great design showcasing that "appropriate technology" designs do not require fashion complexification.

Mind you, in this mixing desk it drives a 47kOhm load (manual value) not 600 Ohm. To turn it into a stand alone "general purpose" Mic-Pre would likely mean to either add additional active output circuit or making changes.

But that is not what this circuit was designed to do, so it's unreasonable to have such expectations.

For a modern design, the discrete OPA may be substituted by a number of different monolithic Op-Amp's, I think the TI OPA828 would make a great substitute. The transformers turn up occasionally in on-line sales and likely are most of what makes the "sound" of the circuit.

Using 2SK2145 for J-Fet & BC850C/BC860C for a modern discrete version using the original design might be fun. Don't forget to avoid loading the output too much, use a buffer to drive 600 Ohm.

Thor

 

[thor.zmt]

I've read most of Mr. Self's work, in my opinion he has very good ideas, and he is brilliant in many ways, but there is no one in the world for Mr. Self except for Mr. Self;

I disagree on brilliant. More plodding, conventional. But his unimaginative plodding ways generate a LOT of useful data.

I think you can still read the exchange over at the diyaudio.com forum.

I used to be active there, but my constant challenging of the "received wisdom" (including the Self-ish kind) and pointing out the emperor was naked was considered "disruptive", so after multiple several week bans I just gave up and let them be.

Thor

 

[user 37518]

Ok, I put it in (schematics attached). I used 2SK209 models for the J-Fet's and BC550/560C models for the BJT's.

The J-Fet's have a bit higher Gm than the originals, the BJT's have lower Beta than the originals so I think the differences will mostly wash out.

Result for open loop gain:
View attachment 110388

The original circuit offers 94dB open loop gain with appx. 1kHz -3dB point (including Mr Miller). Uncompensated we have around 10kHz -3dB. going to 600 Ohm loading drops open loop gain a few dB, nothing major.

Closed loop gain for the M-15B (which I used as it has more details in the service manual) ranges from ~ 6dB to 56dB, so NFB ranges from 38dB minimum to 88dB at minimum gain.

My suggestion of Mosfet VAS with bootstrap load matches the original above 3kHz and increases LF gain (@ 100Hz to 114dB or 20dB more than original).

With a Darlington VAS + additional input stage degeneration (100R per J-Fet) overall gain is lower but LF gain is boosted.

With gain set to 12.7dB (with input transformer this gives ~ 33dB Gain) the closed loop frequency response is identical. Unless stated this setting (gain control @ 50%)

With 300mV output (nominal signal level according to Manual) and 10kHz all three circuits simulate as ~0.0003% THD with H2 being the only observable harmonic above the "noisefloor" of the sim and the original circuit being marginally superior to both modifications.

Raising signal by 20dB (still 10kHz) gives the original circuit 0.0017% THD with H2 & H3 notable.

The Mosfet version (e.g. mine) has 0.0022% THD and the lowest H3 (~3dB less than Darlington/degenerated and > 6dB lower than original), with Darlington inbetween. Going to 600 Ohm loading does degrade THD somewhat, around 3dB or so.

Simulations at 1kHz or 100Hz would likely favour the modified designs, but the value of these advantages may be questionable.

For noise, the darlington VAS is marginally worse than the other two, nothing I'd loose sleep over. The observed 4nV|/Hz @ 1kHz Ein are not that great, however we need to remember that the 1:10 step-up transformer on the input would lower this tenfold.

Unweighted SNR at maximum gain and 0.3V output level looks -60dB. Note that this is at 56dB Gain (actually 76dB total gain including transformer), so that is a helluvalot gain!.

Realistically, the original circuits offers fine performance that gives up very little to modern monolithic Op-Amp's in the way it is used, despite the very simple and unfashionable design. Non of the suggested "improvements", mine included appear to offer any material improvement.

As is, in the original context it is a great design showcasing that "appropriate technology" designs do not require fashion complexification.

Mind you, in this mixing desk it drives a 47kOhm load (manual value) not 600 Ohm. To turn it into a stand alone "general purpose" Mic-Pre would likely mean to either add additional active output circuit or making changes.

But that is not what this circuit was designed to do, so it's unreasonable to have such expectations.

For a modern design, the discrete OPA may be substituted by a number of different monolithic Op-Amp's, I think the TI OPA828 would make a great substitute. The transformers turn up occasionally in on-line sales and likely are most of what makes the "sound" of the circuit.

Using 2SK2145 for J-Fet & BC850C/BC860C for a modern discrete version using the original design might be fun. Don't forget to avoid loading the output too much, use a buffer to drive 600 Ohm.

Thor

I appreciate you took the time to simulate all the scenarios. I find it comforting that our assumptions regarding 94 dB of gain, the 3db knee, etc... are now validated.

Regarding the modifications. Both the Darlington and the MOSFET ones are superior to the original. Also, I noticed that in the MOSFET version you have an inclusive compensation which taps directly from the output of the entire amp rather than from the drain. Plus the extra cap C16. Did you try doing something similar with the Darlington arrangement?

What about stability? the loop gain magnitude and phase.

I was trying to mess with the circuit, but I am afraid I am not well versed regarding Tina, I practically never use it, I rather use LTSpice and I am too lazy to re-draw it in LTSpice.

 

[thor.zmt]

I appreciate you took the time to simulate all the scenarios. I find it comforting that our assumptions regarding 94 dB of gain, the 3db knee, etc... are now validated.

Yup.

Regarding the modifications. Both the Darlington and the MOSFET ones are superior to the original.

Hmmm. In some specific areas and ways, but I have my doubts in practice it will matter.

Also, I noticed that in the MOSFET version you have an inclusive compensation which taps directly from the output of the entire amp rather than from the drain. Plus the extra cap C16.

Inclusive compensation is natural to me, I do it unconsciously.

The extra cap is an old trick in lie of a CCS load. It's a "bootstrap" load which creates a very high load impedance, by "bootstrapping" that load resistor.

Did you try doing something similar with the Darlington arrangement?

No. Nobody suggested anything like that.

I just wanted a quick look if "gilding the lilly" would actually give improvements.

What about stability? the loop gain magnitude and phase.

Can all be run, based on the OLG graph I expected the same stability for Ann versions. But I did not do a rigerous analysis, I am not trying to do a full design.

Just to check if things were as bad as some claimed and how much simple changes may gain, if anything.

Thor

 

[Walter66]

View attachment 110368

I am nostalgic for imperial units. Metric is better for universal use, but...

Thor

A pint is a pint and that will never Change again!
I dislike Barrels, pounds, Inches, pennies. It makes For many Problems, Just try to repair anything Inches in Germoney. Someone once did send me Altec screws, Just Not so easy to fetch in Europe. Great Guy!

 

[Newmarket]

We just all need to adopt metric terminology from now. Even if that means "2.54 mm"

 

[user 37518]

Hmmm. In some specific areas and ways, but I have my doubts in practice it will matter.

In mere terms of mere gain, the low frequencies should have lower THD (probably just a tad less), so as you say, perhaps it is not very significant. But the Darlington and your MOSFET implementation might further reduce THD due to less capacitance modulation or other similar effects.

 

[thor.zmt]

We just all need to adopt metric terminology from now. Even if that means "2.54 mm"

Yes.

And we mustn't use the "Master/Slave" terminology in electronics.

Or the old rhyme to remember resistor band colours:

Teacher who taught pupil racist 'rape' rhyme so he could remember lesson is guilty of misconduct

And black sheep can no longer be black:

Toy maker changes Baa Baa Black Sheep lyrics in "political correctness gone mad"

Now let's not forget to hang up our rainbow flags, wear our rainbow ties and repeat "(only) black lives matter" and go full on woke.

I for one do not think that this woke stalinist purge of our language serves any purpose, but it eliminates and hides history.

Let the language be, leave out yards, feet's and inches and pounds and ounces. Teach people the conversion of units and the history behind the language and why, in our enlightened 21st Century we usually do not use language this way.

Censorship and newspeak simply take us another step closer to "lingua tertii imperii" the wrong way around.

Too many good people died in the 20th century to protect freedom, including that to free speech (including the right to offend with speech), to let it be undermined like that!

End of sermon. And:



Thor

 

[thor.zmt]

In mere terms of mere gain, the low frequencies should have lower THD (probably just a tad less), so as you say, perhaps it is not very significant. But the Darlington and your MOSFET implementation might further reduce THD due to less capacitance modulation or other similar effects.

I think if really want to improve on this circuit, I'd suggest the following:

4 quadrant differential input stage with J-Fets used as buffers

Darlington VAS

Diamond Transistor output

Something like this "less simple mic pre":



Power supply Regulation is intentionally shown as the power supply forms a part of the circuit. Pre-regulation via 3-Pin regulators or a switching DC-DC converter is presumed.

2SK2145 as buffer, MMBT4401/4401 as input transistors, BC847C/857C duals for all else and BCP53/56 as outputs, all SMT. Feedback loop and open loop HF rolloff matches the original, could be optimised more, I am sure.

In the Sim 10kHz THD for 3V into 600 Ohm is 0.00045% and ~0.0002% for 10k load, pure H2. At 100Hz THD is improved 3-6dB depending on load.

Noise is same, so limited by the J-Fet's and resistors methinks. So again room to tweak.

This would be appropriate as a stand alone Mic-Pre. Input Transformer 1:10 is assumed and a 1:1 Line out transformer may be used instead of the impedance balanced output.

Thor

 

[ruffrecords]

Yes.

And we mustn't use the "Master/Slave" terminology in electronics.

I am surprised we are still permitted to use male and female when referring to connectors.

Cheers

Ian

 

[thor.zmt]

I am surprised we are still permitted to use male and female when referring to connectors.

Only for CIS gendered traditionally coloured (non rainbow) connectors.

For those with funky colours you must use:

"Male/Female assigned at manufacturing"



Thor

 

[user 37518]

I think if really want to improve on this circuit, I'd suggest the following:

4 quadrant differential input stage with J-Fets used as buffers

Darlington VAS

Diamond Transistor output

Something like this "less simple mic pre":

View attachment 110450

Power supply Regulation is intentionally shown as the power supply forms a part of the circuit. Pre-regulation via 3-Pin regulators or a switching DC-DC converter is presumed.

2SK2145 as buffer, MMBT4401/4401 as input transistors, BC847C/857C duals for all else and BCP53/56 as outputs, all SMT. Feedback loop and open loop HF rolloff matches the original, could be optimised more, I am sure.

In the Sim 10kHz THD for 3V into 600 Ohm is 0.00045% and ~0.0002% for 10k load, pure H2. At 100Hz THD is improved 3-6dB depending on load.

Noise is same, so limited by the J-Fet's and resistors methinks. So again room to tweak.

This would be appropriate as a stand alone Mic-Pre. Input Transformer 1:10 is assumed and a 1:1 Line out transformer may be used instead of the impedance balanced output.

Thor

Well yes, I would say that is definitely an improvement. It is a completely different beast. I like the diamond buffer but it is not something I routinely use, I should perhaps take it more into consideration. Good job.

 

[thor.zmt]

It is a completely different beast.

Yes and no. It's still 3 Stage LIN, just rail symmetrical and with "folded" emitter followers (aka diamond Transistor or Baxandall Super Pair) which mostly eliminate BJT thermal drift.

I like the diamond buffer but it is not something I routinely use, I should perhaps take it more into consideration. Good job.

It has some advantages over the classic EF2.

FWIW, BTW, as the VAS is heavily degenerated, OLG is only 100dB and uncompensated manages a very wide bandwidth.

It would probably be worth refactoring the whole compensation to take advantage, by using lead/lag over inclusive miller.

Then use a stepped, capacitive compensated gain control.

Thor

 

[user 37518]

Yes and no. It's still 3 Stage LIN

Well, yes. You are right, but its pretty much like saying that a Ferrari and a Volvo are similar because they are cars.

It would probably be worth refactoring the whole compensation to take advantage, by using lead/lag over inclusive miller.

That is one of the advantages of amplifiers such as this one which everything is already set, just like a power amp, not like a DOA which each user might do something different with it. Go for it.

 

[murrayatuptown]

I don't use any archaic colloquial jargon for side- or diagonal- cutters anymore. Not even the L-word.

I never needed the resistor nursery rhyme.Later I got spoiled by alpha-numeric markings on quality resistors. But they can get surprisingly confusing if you find different vintage Mil-Spec resistors with quirky markings intuition doesn't resolve...then have to go find the spec online.

 

[murrayatuptown]

Sometimes ignorance is bliss. I was matching 1% metal film resistors for a project, because I could...had access to a 6-1/2 digit DMM and used 200 line cycle period averaging or similar. The 8-digit meter took too long and seeing all those digits unstable was unnerving, despite knowing I didn't care about them.

What WAS most disturbing was measuring carbon film.with unknown t.c. of R. Air currents made the resistance never settle (closer to the decimal (ok, radix) point. For a little while I thought air never use carbon film again, and got over it. If it doesn't matter, I just read the stripes and don't measure!

 

[thor.zmt]

I don't use any archaic colloquial jargon for side- or diagonal- cutters anymore. Not even the L-word.

I do and I cultivate the use of archaic jargon, though I usually dial down really offensive stuff.

Why? Wokeness has gone too far. It threatens the very base values of our free world. Wokeness in now closer to authoritarian ideologies than to liberal ones.

And Freedoms not excercised disappear.

So I'm:

• pro archaic jargon,
• pro preserving free speech,
• pro gun (regulated)
• pro open carry (not loaded mind you, weapon and ammunition carried separate)
• anti concealed carry (concealed carry should carry 20-life)
• pro offensive speech (that is not hate speech calling for violence against others)
• pro drugs...

Thor

 

[ruffrecords]

I don't use any archaic colloquial jargon for side- or diagonal- cutters anymore. Not even the L-word.

I never needed the resistor nursery rhyme.Later I got spoiled by alpha-numeric markings on quality resistors. But they can get surprisingly confusing if you find different vintage Mil-Spec resistors with quirky markings intuition doesn't resolve...then have to go find the spec online.

I learnt the nursery rhyme in the early 60s but the first word was Bad. It was not until threads like this one appeared nearly 60 years later that it occured to me that some people might think it could be an alternative. I have never heard anyone say or use anything other than Bad.

Cheers

IAn

 

[murrayatuptown]

I'm being (more) absurd (than usual), :0| but if Aretha Franklin had an interest in resistors, she'd have used R.E S.P.E.C.T. for '2' and it wouldn't matter if the '0' Boys were Bad, Bald, Blind...

My first year in an engineering college in the 70's (19, not 18), there were extremely few women students, and they were WAY nerdier than the men.

Decades later, the women graduating with engineering degrees seem quite normal (more or less). Or they're Civil Engineers, or went to a different school (!).

My daughter has a female friend who is an engineer and met her husband in college. Regarding that, she said 'the odds weren't good, and the goods were odd'. (Duality principle applies)

 

[user 37518]

I'm being (more) absurd (than usual), :0| but if Aretha Franklin had an interest in resistors, she'd have used R.E S.P.E.C.T. for '2' and it wouldn't matter if the '0' Boys were Bad, Bald, Blind...

My first year in an engineering college in the 70's (19, not 18), there were extremely few women students, and they were WAY nerdier than the men.

Decades later, the women graduating with engineering degrees seem quite normal (more or less). Or they're Civil Engineers, or went to a different school (!).

My daughter has a female friend who is an engineer and met her husband in college. Regarding that, she said 'the odds weren't good, and the goods were odd'. (Duality principle applies)

At least from my perspective (been teaching EE for years), things haven't changed much. Women are still very scarce, usually they are nerdy, shy, well behaved, and polite.

 

[user 37518]

I'm being (more) absurd (than usual), :0| but if Aretha Franklin had an interest in resistors, she'd have used R.E S.P.E.C.T. for '2' and it wouldn't matter if the '0' Boys were Bad, Bald, Blind...

My first year in an engineering college in the 70's (19, not 18), there were extremely few women students, and they were WAY nerdier than the men.

Decades later, the women graduating with engineering degrees seem quite normal (more or less). Or they're Civil Engineers, or went to a different school (!).

My daughter has a female friend who is an engineer and met her husband in college. Regarding that, she said 'the odds weren't good, and the goods were odd'. (Duality principle applies)

At least from my perspective (been teaching EE for years), things haven't changed much. Women are still very scarce, usually they are nerdy, well behaved, and polite. Men on the other hand....

BTW I can always tell if something by my students was handwritten by a man or a woman. Women do it much better and looks nicer.

 

[murrayatuptown]

I tell people who complain about MY handwriting I was supposed to grow up to be a doctor.

Or if they bring me something and ask if It's my writing I say 'if you can read it, it's not mine!:.

 

[murrayatuptown]

I had a salesperson from Taiwan visit a factory I worked at, and she remarked that I wrote very well for an engineer. I admitted/joked my secret was that I was a drop-out engineer, not a pedigreed one, and complimented her writing skills in a secondary language.

Her secrets were that she had a psychology degree and was also the company president. She was either an only child or had no brother, so her father expected her to take the helm. (My uncertainty, not hers...it was 1995).

IIRC, the late Jim Williams of what used to be Linear Technology, had a psychology degree, but also a profound level of interest and skill in what became his career. I think he repaired Tektronix gear along the way.

 

[ct.waveform]

I didn't know that about Jim Williams. I'd always put his brilliance down to having a single syllable Christian name. All the greats had/have just one syllable; Jim, Bob, Walt...

 

[user 37518]

I had a salesperson from Taiwan visit a factory I worked at, and she remarked that I wrote very well for an engineer. I admitted/joked my secret was that I was a drop-out engineer, not a pedigreed one, and complimented her writing skills in a secondary language.

Her secrets were that she had a psychology degree and was also the company president. She was either an only child or had no brother, so her father expected her to take the helm. (My uncertainty, not hers...it was 1995).

IIRC, the late Jim Williams of what used to be Linear Technology, had a psychology degree, but also a profound level of interest and skill in what became his career. I think he repaired Tektronix gear along the way.

I don't think Jim Williams had any degree whatsoever. He was one of a kind. Bob Pease only had a BSc EE from MIT, no master's, PhD, whatever.... Also, IIRC, Widlar was also a BSc, no master's or PhD, although he might have had an Honoris Causa Doctorate in binge drinking.

 

[ct.waveform]

I don't think Jim Williams had any degree whatsoever. He was one of a kind. Bob Pease only had a BSc EE from MIT, no master's, PhD, whatever.... Also, IIRC, Widlar was also a BSc, no master's or PhD, although he might have had an Honoris Causa Doctorate in binge drinking.

You've named the highlights of my list of greats with single syllable names. I should have included Bill, for Bill Whitlock, who has an unmatched understanding of those all-important Grounds. He deserves inclusion just on the strength of his amusing observation that solving ground loop hum invariably ends with the fantastically unscientific "That's good enough." or "That'll do.". After all the work that has gone into SNR, that's how it usually ends.

 

[Newmarket]

Of course, ,most names can be shortened to one syllable if you want to...
now I'm trying to think of exception...
Jesus ?
Harold ?
Mary ?