Need a Discrete Transistor preamp stage (for summing amp)

Can anyone recommend a preamp output driver stage schematic composed of discrete transistors (instead of opamp ICs).

single supply voltage... 18V (MUST!)

unbalanced or balanced (pref. balanced)

No output transformer preferably (but may go with transformer-based design too... hopefull not expensive trafo)

I'm "upgrading" an old Altec mixer. The output stage will be used as a summing amplifier (tie together 6 channels via resistors) and/or output driver.

Do you know of any schematic I can "borrow" bits and pieces of the design? Thanks!

you could do an API 325 type card, with melcor opamps, and pickup the 2503 output traffos from tommy for roughly $20 (US) a pop.


I guess this is akin to what kubi just said :grin:

ju

Yeah I have that first stage of the Neve preamp with the 3 transistors... but it's 24V. Do you think it will work with 18V?

Of course, I can regulate 24V and 18V on my power supply....

> instead of opamp ICs

What is wrong with ICs?

> output driver stage schematic composed of discrete transistors... single supply voltage... 18V (MUST!)

So what are the rest of the specs? Do you need gain? How much? What maximum output voltage? What is the load impedance? And what is the source impedance? If you have one of the low-parts-count Altec tube mixers, the preamp outputs may be 100K-470K impedance, which has a major influence on design.

This minimalist plan will give hi-fi / semi-pro / Tascam level outputs:

Thanks PRR.

Nothing wrong with ICs, but the power supply is single-ended (not split) so I don't want to use any opamp needing +/-18V. I'm currently testing an unbalanced to balanced driver using LM324 (single supply) opamps.

Also I don't want to diverge from the "vintage" look and feel of the Altec, thus my requirement for using discrete transistors.

I'm thinking using discrete transistors, and leave the output unbalanced, or continue some more and add the opamp balanced driver stage.

I found a ShureM68 (which is kinda the same configuration as my Altec). This is their summing amp/driver circuit.
http://www.fivefish.net/diy/Altec1599A/outputdriver.gif

ummm, nevermind =)

Take a decent op-amp. Dependent on the number of inputs connected, ensure it is stable for the gain you require (10 inputs to a virtual earth mixing amp set for unity gain, is equivalent to x10 gain...)

Connect power rails to 0V and +18V. Bias the +ve input to mid rail with a potential divider consisting of two equal resistors (one up to rail, one down to ground). Remove any power-rail ripple with an electolytic cap from +ve input of the opamp to ground (say with 10k resistors, 47uF, half rail rated should be loads - no signal goes here).

Install a negative feedback resistor from the output of the opamp, to the -ve (inverting) input. The value of this should be the same (usually) as the mixing resistors.

Put a big cap on the outside of this, +ve facing the -ve input of the opamp. Connect the -ve of the cap to the mixing bus. How big? If N is the number of channels being mixed, and R is the value of the mixing resistors, then C (in microfarads) should be conveniently bigger than:-

(1000000*N)/(30*R) (for a 5Hz LF turnover)

Decouple the DC from the opamp output with a decently large cap (47uF will probably do), +ve toward the opamp output, stick a 100k resistor from the -ve side of the cap to ground for decencies sake, and Robert is your jolly old mother's brother...

Hope this helps. (A picture paints a thousand words - sorry I don't have a convenient host on which to stick a diagram - I'm sure someone else does.)

If you want a discrete equivalent, that is certainly possible, but someone is going to have to *design* it ... such is the attraction of the wonderful opamp...

[quote author="AP"]

Hope this helps. (A picture paints a thousand words - sorry I don't have a convenient host on which to stick a diagram - I'm sure someone else does.)

[/quote]

Hi AP,
You can use

http://www.villagephotos.com/

with ANY email address poped up in your head, no confirmation needed! :thumb:

BTW, you work for marshall? :green:

*laugh* :grin:

BTW, you work for marshall?

Click to expand...

No no no :razz: - their factory is about 200 yards from where I work though. I have not been in the audio business for maybe 25 years, but I was (for my sins) one of the founders of Studiomaster (ie. the engineer).

Alan

> LM324 (single supply) opamps

324 is (for most circuits) an awful audio amplifier. Even I can hear it sucks chunks. By the time you fix the worst problem, any other chip would be a better plan.

> I don't want to use any opamp needing +/-18V.

As Alan says: essentially any op-amp can be run single-supply. IC op-amps don't know where "ground" is, and are not fussy about supply voltage.

Here is the 2-cent design:


Since your supply voltage (and max output voltage) is low, and you wanted balanced, I show push-pull output.

Op-amps can be anything from 5532 to TL072 (please, no 324). At these impedances (as from a simple tube mike amp), the TL072 may be as low-noise as the 5532.

For another design penny, I'd suggest you throw 47-100 ohm series output resistors on there, maybe 5K to ground to drain leakage from the output cap. You might need a little resistance (1K-5K) between the mix amp and a large summing bus to avoid RF instability, but that is where a designer earns the big bucks (fixing obscure real-life problems).

[quote author="AP"]

BTW, you work for marshall?

Click to expand...

No no no :razz: - their factory is about 200 yards from where I work though. I have not been in the audio business for maybe 25 years, but I was (for my sins) one of the founders of Studiomaster (ie. the engineer).

Alan[/quote]
Ah... so *You're* responsible then!!!

:wink:

A friend of mine had (...lord knows how many!) RSD/Studiomaster consoles back in the 1980's... he never learned! :green:

Keith

Thanks PRR.

I think I'll use NE5532 chips using that push-pull circuitry with the offset voltage on the (+) input.

Okay so essentially my new plan now is using (4) NE5532 chips (8 opamps total).

6 opamps will be used as additional gain for each channel (so I can do (6) direct outs), and the other 2 opamps as balanced driver in the summing stage.

BTW, is there a Quad NE5532 opamp?

Ah... so *You're* responsible then!!!

Click to expand...

... Err - not me guvnor! I was gone as of 1978... (Although I did a bit of a comeback a few years afterwards, and designed a desk called the P7)...

... Story! A masterpiece of imaginative marketeering. This design was our 7th audio design project at that time. The Studiomaster marketing minds cogitated at great length for a name for the new desk. Eventually, they called it P7. (Every drawing and schematic had a number that started P7, you see - Project 7).

Oh well, you had to be there, I guess...

I think I'll use NE5532 chips using that push-pull circuitry with the offset voltage on the (+) input.

Click to expand...

... err - technically termed the "bias" voltage...

Okay so essentially my new plan now is using (4) NE5532 chips (8 opamps total).

Click to expand...

It is not unknown to have an additional level control - a master fader - after the group mixing bus amp. You might consider that if you have somewhere to physically put it.

... So the mixing amp will drive the fader (that will look just like the top op-amp alone in PRRs rather artistic diagram), then the wiper of the fader can be buffered by the line drivers. (Just like his two opamps all over again, but with a 33k resistor from the wiper to the -ve input of the top opamp (you need to keep the blocking capacitor though - this will give you 10dB of post-fader gain, should you ever need or want it).

Alan

Did Studiomaster make some crap consoles in their later days?

I was under the impression that their stuff was regarded as a reasonably-priced but solid performer on the second-hand market.

Did Studiomaster make some crap consoles in their later days?

I was under the impression that their stuff was regarded as a reasonably-priced but solid performer on the second-hand market.

Click to expand...

Oh no - I'm sure you are right - it was pretty good value for money all the time. Quite a lot of functionality for not too many bucks.... but it was just not a Neve!

(I was just donning a steel hard-hat, in case someone here thinks otherwise....)

Alan

If it makes you feel any better, I've never heard of StudioMaster before this thread. :thumb:

Well, okay, maybe that won't make you feel better... :roll: :green:

I have (stored in my garage) a Studiomaster 16X4 from the late 70s. For a mixer made from 741-type opamps, it was pretty okay, I thought.

Is that one of yours, Alan?

Is that one of yours, Alan?

Click to expand...

... guilty as charged, I'm afraid.

Well, even given the 741's, it was pretty good for its day - nothing really like it at the time - transformerless balanced input, absolutely identical to this:-

http://sound.westhost.com/project66.htm (I wonder where that came from?)

Continuously variable frequency bass & mid. All ready for recording use with the 4-track 1/4" Teac recorders that had just become available. Also, it was none of the first serious non-modular (read: affordable) desks... I was pretty happy with it at the time...

PS. Circuitry was somewhat similar to the Mixdown & 16/2 schematics here...

http://mysite.wanadoo-members.co.uk/studiomaster/

Alan