Bauman Neve output in a PRR varimu..

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
it's making some sense but I am unfortunately not a sidechain guru. It's at least a good starting point for discussion if not a starting point to get people motivated.

:thumb:

EDIT: I have another PCB that I had made when I built the first version of your Varimu that I could use for some of the prototyping when we decide to move forward.. I also found a Ratshack with some of the old filters too.. I suppose i should go ahead and pick those up.
 
To summarize: Svart wants to try my 12AU7-based limiter, but with a Neve BA283-style output stage instead of a Tascam-class 5532 -10dBV output.

12AU7 limiter - - BA283 info

The original seems to be a good balance between noise and distortion{edit}, so I don't want to change the levels in the 12AU7{/edit}. For nominal +4dBm reference, it needs 4 times the output level of a -10dBV system, {edit} so we raise the output stage gain{/edit}. The original had output gain about 10, we need output gain of 40. Gain of a BA283 is complicated by output-iron step-up and internal resistors: the strapping for Gv=40 to the output seems to be 150Ω and 100uFd from pin K to signal ground. The BA283 also has an aux unbalanced output, which we will tap for the sidechain; it runs 4dB lower level than the transformer output.

The 12AU7 wants about 100 volts, the BA283 needs ~24V, the 12AU7 needs 12V heat, and my original output amp and sidechain ate +/-16V with a 1.5V bias. What a mess! And technology from every era, like an old-time scrapyard. Can this be cleaner?

How would Neve have approached this? His companies made a lot of spiffy discrete single-supply circuits, some extremely elegant. How to slam a 12AU7's grids to -5V using only the one +24V rail already needed for the BA283 (and if we run stereo, 2*12AU7 heaters)?

After much wild goose, I came to the obvious idea: hold the 12AU7 cathode up at +12V or so. Then the grid resistor (the limiter's release resistor) pulls the grids up to +12V for full gain, the sidechain pulls it down to ~+7V for minimum gain.

12AU7-BA283.gif


I stuck a 12V Zener under the 12AU7 cathode. The 12AU7 current may fall very low at high gain reduction, so it may need a bleeder, such as 5K to the +24V rail, to keep Zener current from falling too low.

We need a peak||peak rectifier. It should detect asymmetrical waveforms correctly. There are a zillion rectifier designs, a dozen that will work correctly. I'm showing one that is "new to me", though probably not original.

Q2 Q3 only turn on when they get more than 0.7V at their Bases. They are driven from phase-splitter Q1, so peaks over 0.7V of either polarity will turn-on Q2 or Q3. When Q2 or Q3 turn-on, current is pumped into the 20K-100K Ratio pot. At idle it sits at +12V, current drops this point toward ground. How fast? The ratio of the Ratio pot value to the 4K resistor in Q2 Q3 Emitters.

Anybody see why this won't work?

We could put a time-constant cap here, but the impedance is variable and rather high. Q4 buffers this (and also helps ignore slight leakage in the Ratio pot). The 50K-1Meg Release resistor returns to +12V. The 1K-50K Attack resistor feeds the 1uFd time-constant capacitor. This cap returns to ground so transient currents don't get injected into the 12AU7. The voltage on the cap feeds the 12AU7 grids, either via a center-tapped winding or through two fairly-small resistors.

GR metering can sense either 12AU7 cathode current or grid voltage. This is a little annoying because both points are referenced to +12V. This complicates using one meter for either output level or GR level. But you have level meters on your recorder. A milliampmeter plus a shunt resistor in R3's location is the simplest way to indicate relative GR. Adjust the shunt so the meter indicates full-scale at no-signal.
 
wow, that's a great start! I'll be digesting this over the holidays.

Are we concerned with any certain specs on Q1-4? It seems that they need not be anything out of the ordinary from a quick glance... I'm thinking BC550 for Q1-3 and BC560 for Q4 since a lot of people would have these around from other projects..

As for the power supply, We can get very close to 24VDC with an 18v trafo, around 25.2VDC.. this should cut down on the wasted heat a bit. A quick look shows TE62081-ND(digikey) 50VA, 18v in series @2.7A for 18.83$. Likely overkill but only a buck or so more than the >50VA.

Gotta get back to work now but I'll be thinking this over more today.

Thanks PRR! :thumb:
 
PRR, are you sure that this will work? Q2 and Q3 are fed by an inverted phase. If one opens, the other closes. And then it will be summed together and any change is chanceled out.
 
> are you sure that this will work?

That's why I posted: to see if anybody could find fault.

> summed together and any change is canceled out

I do not think so, though I am not sure.

Q2 Q3 are both normally off. Large peaks turn one or the other on, maybe both but not at the same time. Their input is AC but their output is unidirectional pulses. I think.
 
> any certain specs on Q1-4? It seems that they need not be anything out of the ordinary

20V, 100mA, and probably Hfe over 100, though >250 may be better. Probably "anything" will work, though I may have to revise the Q1 stage.

> As for the power supply, We can get very close to 24VDC with an 18v trafo, around 25.2VDC

The BA283 was designed for batteries. Ideally, trickle-charged batteries, but it was expected to keep working through local utility failure. (It is broadcast-tradition, not yet a recording-only tradition.) So yes, it will work 22v to 28V. Still, I have real doubt about feeding it varying voltage in non-emergency work, especially in light of buttachunk's recent low/snow-voltage sessions. Its sound will change with supply voltage. Not a lot, but maybe enough to notice. I'd rather have it consistent.

> this should cut down on the wasted heat a bit.

Waste heat? Let's draw a stereo-limiter budget:

12AU7 heaters: 6.3V @ 0.6A ==== 3.8 Watts
12AU7 plates: 100V @ 0.012A === 1.2 Watts
12AU7 regulator: 60V @ 0.012A == 0.7 Watts
BA283s: 24V @ 0.2A ========== 4.8 Watts
BA283s regulator: 12V @ 0.2A ==== 2.4 Watts
sidechain: ==================== 0.1 Watts
TOTAL ========================= 13.0 Watts

The BA238's regulator, 2.4 Watts, is not a major part of the 13 Watts. Take it away and you are "down to" 10.6 Watts. Not a big deal.

The 12AU7 plate voltage does need to be pretty stable, because it has a large effect on threshold and ratio. It may not need to be Zener stable, but should not drift with utility company whims.

At the moment, sims say the drive to Q2 Q3 is unbalanced, enough to be an issue. I'm trying to see why. The cathodyne has unequal output impedances; does not look like enough to cause ~30% unbalance but maybe with cap-charging in spike-waves, it is enough. The obvious alternatives mostly replace Q1 with two transistors.
 
interesting!

In wasted heat I simply meant using a higher voltage (PSU) trafo than truly needed and having to either resistively or actively limit the voltage to the various stages.. of course either way we look at it it's going out as heat. 13 watts isn't *that* much but inside a box without active ventilation and without much in the way of heatsinking that could be pretty warm to the touch if not allowing any Vregs(I rather like using them even when i don't *have* to.. :green: ) to likely overheat.

but maybe that is just me being overly sensitive.. I have to worry about junk like that in the daily grind so forgive me!


zener stable.. hey they are cheap and work good so I say jellybean them.

just by looking at it seems that Q2/3 would conduct with a balanced *like* action under an AC signal and both would be off with no signal as you suggested however I am spying the sourcing 4k7..maybe a little lower value? i would even be crazy enough to stick some Schottkys in there for the coupling caps to see what would happen or maybe i'm just too tired to think.. :shock:

or just stick a mosfet in there..that old BJT technology is out to pasture! :green:
 
> cut down on the wasted heat

There IS a non-issue here. My threshold level is one silicon junction, and thus temperature sensitive. I am assuming "shirtsleeve" environment: varying no more than +/-10 degrees C. Over that range, we get a +/-22mV variation in a ~600mV threshold, 578mV to 622mV. This is a +/-0.32dB variation in theoretical threshold. Few studios work to that precision. If they do, they probably would not want to rely on a sloppy 12AU7 scheme. Because it is not a brick-wall limiter, a few-dB change in peak input levels will cause more than +/-0.3dB change in peak output levels, making thermal drift just irrelevant.

If NYDave needs something to put in the company attic, where he can't work more than a minute before heat castration, this may not be ideal. A +50 deg C rise would cause a -1.6dB change in output, which might confound the monthly testing.

However, we have a tube heater plus a Class A 2N3055 in the box. The threshold transistors Q2 Q3 should be well away from heat sources. If this rig were lashed too tight, even setting levels by ear, there could be enough drift from a 5-minute warm-up over the next hour to slightly upset your limiting action.

> just stick a mosfet in there..that old BJT technology is out to pasture!

In the words of a notorious experimenter: "I reject your reality and substitute my own." Yes, FETs are wonderful things. If you insist, use one at Q1 and fiddle the gate bias to put about 6V at the cathode (whatever you call it). Won't raise the input impedance unless you find a very low-C MOSFET. A jellybean JFET will be low-C but may have low enough Gm to lose gain (yeah, it is only unity gain but it can't be much less than unity).

In the Q2 Q3 Q4 strip: FETs won't reduce parts-count, don't have a law-of-nature turn-on voltage, don't have the high Gm of a BJT.

And the real reason: Neve didn't have MOSFETs, and didn't use JFETs. I'd like to keep it all of similar vintage, even in the places that don't "sound". (Yeah, that could mean another 2N3055 to regulate the +24V, and using good British transformers on each side of the 12AU7....)

I've nearly melted a 3GHz Pentium flogging a simulator. I've proved that 500,000 transistors can't keep up with four transistors (to sim a millisecond of audio takes the Pentium over a second, with much lower accuracy than 3906 jellybeans would really give). I am reminded how complex non-linear systems can be. In the end, I think my design above works with a minor change (resistors between phase-splitter and Q2 Q3 to reduce bias-shift effects), but I reduced most impedances to speed-up the attack. It is now in the speed class of a Fairchild, it can catch any likely audio peak. I don't think it needs to be anywhere near that fast on systems that clip politely. I think a 1K Attack resistor, ~1mS, is plenty fast for music into digital recorders or well-behaved speaker-amps, but you can reduce the attack resistor to zero and get near 50uS. Without melting Q4. (Using 12AU7 at 100V, instead of a low-Mu tube at 240V, makes a 10:1 difference in voltage and 100:1 difference in power, which is why a jellybean can do what it took two 6V6 to do.)

Here is the chalkboard tonight.
12AU7-BA283-SC2.gif

The other sketch was right-to-left, as sidechains are often drawn in context; this is left-to-right because I was considering it alone.

The V1 V2 circles are sim sources; in real life an unbalanced main output must be fed through a pot to C1.

For stereo, that pot must be 2-gang, and everything is duplicate up to the "Other_Channel" point. The stuff around Q4 is common to both channels.

I thought it was clever to derive the +12V from the 12AU7's cathode current. That may work but may crap-out at high gain-reduction when we starve the 12AU7. More to the point, it may be very slow to start-up from power-on. Run a 1K resistor from the +24V supply to a 12V Zener to ground. The +12V side of the Zener accepts 12AU7 cathode current and feeds Q4.

The Ratio pot may or may not be useful. The Attack pot should go over 50K for slow attack, to 1K for audibly instant attack (and towards zero if you must NOT ever overshoot, as in disk cutting). The Release pot can run 1Meg for slow to 50K for fast. If all the times are too fast or too slow, change C4.

C1, C2, C3 are electrolytic: assuming input is ground referenced, the + side of the caps faces Q1.

All resistor values may be replaced with handy 10% values (4K7 for "5K"). The R5 R6 R11 R12 set should be matched better than 10% for 1dB symmetry on + and - peaks. For stereo, the two R7s should be matched better than 10%. All the Q2 Q3 parts should be from the same box.

Wild idea to ponder: if you go to this much trouble, I suspect that 6BC8 will be better in some ways. You can't steal a 6BC8 from the nearest guitar amp, the way you can a 12AU7, but it is hardly expensive (and always good USA/EUR NOS). About $5. It needs similar plate voltage at a little more current, it needs a little more heater power with different pin connections, and it needs 6V which is not as serendipitous as two 12AU7 with a +24VDC supply nearby. But distortion and noise could be less, gain a bit more, limiting action a little firmer for the same sidechain response.
 
I am still getting around to this project PRR and those interested. I have parts coming in this week for part 1 of the project, just parallelling a 283 output section with the 5532 stage of the established Varimu design.

I have another PCB etched and ready for major hacking in part 2, the prototype of the suggested Class A discrete version. this should be interesting!

Part 3 will include final revisions and a PCB layout if PRR doesn't get tired of the questions that I have started to jot down.

:thumb:
 
Hi Svart,
I'm doing one of these right now. your part 1 version... pretty much just the straightforward project, based on kent's etch layout with some ba283s for output.

I'm messing with a few things, like the stereo linking on the sidechain.
mosstly just switching problems...

I also have this done as in Kent's board... a 22Ohm and 100r trimmer on the tube bias with the middle leg of the pot feeding my meter...
hmmm, I only want to use 1 meter. can the left and right meter feeds be summed together. will this adversely affect the gain reduction, should I add a buffer?

got to figure this out. not sure if I'd need a dc summing amplifier like so
http://www.aldinc.com/pdf/amp_27002.0.pdf
or just a couple of resistors...

I'd love to hear some thoughts on this. probably help me figure out more about the ultimate ssl sidechain summing... not quite there on that one either.

Thanks
Kelly
 
Hey Tommy, don't know about Svart, but it's pretty funny you posted here today, I was just looking at this post to see if I could figure out how to wire/sum my meter since I'm just about finished with this. I completely forgot my last post... more than a year later and I've picked it back up. I should be able to let you know in a few days how it works.
Sleeper
 
> based on kent's etch layout

If that's the one I think it is: get the heater leads off the PCB, it injects ripple all along the "ground".

> I only want to use 1 meter. can the left and right meter feeds be summed together.

Is this supposed to be "Stereo"?

Then the first goal (after verifying separate operation and trimming thump) is to balance the two channels. Feed the same signal to both. Adjust your monitor balance so the "stereo image" is a thin line down the center, exact Mono. Beat the limiter. If the image shifts left or right, change a tube. Repeat until image shift is tolerable.

You could just watch stereo meters on recorder. But even 1dB of image shift is audible to my ears, and I'm not a fussy guy.

If the image-shift is small, then there is no point in having two GR meters, right? They would always read the same!

Note that separate cathode meters will NOT prove stereo balance. GR is not an exact function of cathode current. It is a function of a lot of tube parameters, which always vary +/-20%, and this scheme pushes well below the currents that the 12AU7 factory tested at.

We meter cathode current because it is simple and convenient, the signal can be run long distances if desired, and there is no better method. Cathode current metering is cheap, simple, dances fine.

With two different 12AU7s, you could have zero image-shift with un-equal cathode currents. If that balances good, who cares what the currents are?

You can't know the actual GR without an exact-match part (some JFET limiters try it this way, trimming an extra JFET "gain reducing" a DC voltage) or an ultrasonic pilot tone (ugh!).

> not sure if I'd need a dc summing amplifier like so http://www.aldinc.com/pdf/amp_27002.0.pdf

It is two currents. They both go to the same place. Just sum them, and use a higher-current meter (actually a 1mA meter with a smaller shunt).

You "could" meter just one. You already trust your stereo image-shift balance better than the rude GR calibrations. However that is likely to throw about 0.1V more drop in one side than the other.

So bring both 22Rs together, through meter, to ground.
 
Sleeper, excellent! PRR, thanks!

I don't have a board, so would build P to P.

Are you actually doing the Neve output stage?
 
Hi PRR,
thanks for the reply.
as always, now not only do I know what to do about my problem but I know more intimately how the thing works...
I'll for sure rewire the heaters, easy and seems like a good call.

I also put a small break in Kent's ground path, which is essentially a big loop around the entire board.

and yes it's stereo. I have matched resistors for each step of the attack and release knobs. the sidechain can be switched for either independant sidechains or summed sidechain. connected just after the rectifiers.

Tommy,
Yeah I'm doing the NEVE output... I had the transformers and the board from some other project... turns out my electrodyne summing amp kicks tons of ass, so I never used the neves. I'll try em here.

Anyhow, let me get this working and I'll let you know what i think of it.
Kelly
 
Here is the project page in ZIP form.

You MUST Save the file, take the .TXT off the end of the file-name, then it should un-ZIP.

(This to work-around the forum's file-type policy.)
 

Attachments

  • Comp5.zip.TXT
    50.2 KB · Views: 38
We never got to know if the result sounded good...

Question for the experts here: is it possible to use the neve output stage without the transformer? (Making it unbalanced)
 

Latest posts

Back
Top