PRR Varimu Mod

So here is a schematic of the things I did to the PRR varimu.

I used the original circuit board layout here:
http://www.diyfactory.com/projects/diy12au7comp/diy12au7comp.htm

when I first built the stock 12AU7 version it sounded nice but I wanted it to do a little more so I started copying some things from the UA175 which used the 6BC8.
PRR suggested using the 6ES8 early on.

So I changed the plate resistors to about 4K and found out the interstage TX for the 175 was an A19. I figured 15K on the plates was a good idea.

I don't know if the 25K dual pot in front of the input transformers is the best place for attenuation but it works for me. someone might want to try it after the input transformer wipers into the tube grid.

Another improvement would be NOT to use the shunt method of volume control for the 990 opamp. A 100K pot could be used instead of the 100K resistor in front of the 990 for a wider range of volume control.

I ran the 6ES8 heaters in series (cut a few traces on the board) so I could still use 12VDC.

I ran the 990 from the same +/- 15 volt power supply used for the other chips. Cinemag input and Triad interstage transformers.

I used a vintage 1ma meter that had VU graphics on it.

6BC8 will also work in this circuit.

Thanks for sharing this.
I've never managed to make my PRR varimu sounding good yet.

I beliive I should mod it like yours.

Just waiting for your samples :thumb: :thumb:

Heu... forgot to ask.

Would a Edcor XSM 15K:600 be nice for the interstage TX ?

I suppose so, cause it seems that XSM series has been used with success within tube circuits.

Any suggestions concerning TX substitute in the circuit above would be nice :wink:

Look good :thumb:
You can try the jensen offset compensation for reducing DC offset at 990 output (as seen in the John Hardy PDF)
Also maybe use 2 * 5534 ICs rather than 1 * 5532 IC for reducing crosstalk ?
And if you use an OPA604 for U3 and an OPA2604 for U6 & U7, you could then power all the ICs with 24V, but I don't know if it will sound better though

I linked a drum sample in the first post. This thing is great on guitars too. it makes my POD XT sound like real recorded guitars. :grin:

really interesting sound. very soft sounding but made the snare pop nicely without drastically raising the hi-hat and cymbals as well. Just curious, but what did we hear here? Was that an overall drum mix that was bussed through the compressor? How many mics were included? The only thing I did not like was that it brought up the residual noise of what I think is the floor tom when it was not being played (hearing it vibrate with each kick hit).

I linked a drum sample in the first post. This thing is great on guitars too. it makes my POD XT sound like real recorded guitars. Very Happy

Click to expand...

put something like that on gearslutz or andy sneap forum and there will be a lot of new PRR vary mu :twisted: .

did you do a compression curve plot? or at least how many dB can it compress?

The drum sample was a multi miked set. They're tracks from a record I'm working on. They're dry with no EQ. just the tracks quickly level mixed and bussed through PRR varimu and back in to the computer. The PRR was on pretty fast attack and release.


I have no plots or graphs. I don't think the meter is that accurate either. But I just love what it does to all kinds of things. Sounds great on Piano.

Oh, and I'm sure the Edcor transformers would work great. I was using a UTC ouncer for a while and did just fine. I later replaced it with a Triad the size of an A series UTC.

You should put a capacitor in the feedback loop of the 990 for stability. Possibly around 47pF or so.

Also, as drawn, you could accidentally adjust the 10k pot at the inverting input of the 990 to zero ohms, resulting in infinite gain from the 990 (ignoring the effect of the series capacitor on very low frequencies/DC). You need to add a resistor in series with the pot to limit this, based on the maximum gain you want from the circuit.

Another improvement would be NOT to use the shunt method of volume control for the 990 opamp. A 100K pot could be used instead of the 100K resistor in front of the 990 for a wider range of volume control.

Click to expand...

If you were to use a volume control in place of the 100k resistor, I would use a lower value than the 100k that you suggested. This would provide a lower overall source impedance for the non-inverting input of the 990, resulting in lower noise from the 990. A 10k pot or lower would be better, although it will have some loading effect on the circuit around U3.

Other than that, there are obviously other things that could be done to customize the circuit, but I'm mainly concerned with the application of the 990.

Best of luck.

John Hardy
The John Hardy Co.
www.johnhardyco.com

Oops

Yes there is a 47pf (or it might be a 100pf) in the circuit across the 20K.

Also all pots are on rotary switiches so the 10K never gets to zero ohms. I think I have a 500 ohm to ground on the last step of the volume switch.

I'll update the schematic...

Thanks for catching that.. :shock:

If you have any other suggestions or improvements I'd love to hear them.

Bluebird, thanks.

John, did you have other suggestions? I don't think you'll step on any toes here if you have some ideas to suggest.

John, did you have other suggestions? I don't think you'll step on any toes here if you have some ideas to suggest.

Click to expand...

I'd rather stick to the 990 and transformer-related issues at this point. I'd have to study the rest of the circuit, and I'm swamped. I appreciate the thought though.

My only other concern would be to check the optimum way to use the various transformers. In my M-1, M-2 and Jensen Twin Servo mic preamps, I use the JT-11-BMQ output transformer (basically the same as the JT-11-BMCF) with the Jensen JT-OLI-3 output isolator in series with the output of the 990 on its way to the high side of the primary of the transformer. The JT-OLI-3 consists of a 39 ohm resistor and a 3.7uH inductor (wound on the body of the resistor) in parallel with each other. The JT-OLI-3 provides essentially zero-ohms impedance in the audio bandwidth, and 39 ohms above the audio bandwidth to provide maximum signal transfer in the audio bandwidth while still providing isolation from capacitance beyond the audio bandwidth.

Actually, there is a 2nd JT-OLI-3 that goes from the low side of the primary of the transformer to ground to provide a balanced output impedance for optimum common mode rejection. Same impedance conditions at the high side and low side.

Some line-output transformers perform better when they see a zero-ohm output impedance from the op-amp that drives them. The 47 ohm resistor in the compressor schematic would increase the distortion of some transformers. But, it depends on the transformer.

There may be issues with the dual 25k pot at the input in terms of how it affects the frequency response of the input transformer, and the CMRR of that transformer at different pot settings due to the changing impedance that the transformer sees, and the imbalance of the two pot sections due to relatively poor tracking. In any case, put a 1kHz square wave through the circuit and see how it looks when it comes out during various gain settings, etc.

John Hardy
The John Hardy Co.
www.johnhardyco.com

bluebird

did you try edcors with your revised schem? id love to do this comp with edcor xsm's

No I haven't tried Edcors but like I said I've tried two different kinds of 15K to 600 transformers and they both worked great. So Yea, Edcors would work just as well I'm sure.

yea the input pot always bugged me but since 90% of the time I have it turned up all the way, the mismatch of the pots never seem to come into play.

if you look at the UA175:
http://www.waltzingbear.com/Schematics/Urei/UA_175.htm

They have a pad type thing after the input transformer. This type of switch looks like it would lower the input signal while giving the grids more negative control voltage.

What WOULD work well would be this type of switch in front of the transformers with 1% resistors. Just something to knock down a real hot signal. Thats all Ive ever had to do is just turn down the input a bit on some signals. It could be made with a simple DPDT switch.

> the interstage TX for the 175 was an A19.

So why does your hack-plan show "600" on the right side?

The A19 was sold as 15K plate to 80K push-pull grids. In the 175 it is used backward, 80K:15K. Not clear which way you have it. Or what input tranny you have.

The 80K side probably does not show the 990 module to best advantage. On the source material you choose, at the depth you used in the sample, this may be unimportant.

????

Oh I didn't know the A-19 was used backwards.

I'll just chalk it up to a happy accident then.

Now that I think about it, it would make sense if 80K to 15K is a step down. 15K to 600 is also a step down to the next stage. The plates of the 6ES8 would be higher impedance than the input of the 990 right?

Like I said before I'm sure there could be a lot of improvements but I just wanted to share this "hack" version because I thought it sounded great like pictured in the schematic.

As far as effort, if building from scratch you might want to try Analag's new Pm660.

BUT if you have a prr already you might want to try a couple of these changes. you could even use a 5534 instead of a 990.

:grin:

any chance of getting that schematic back up? - Thanks, Ben

yes please!! missed this one too. Mod Schematic would be great!
cheers
M

me too

Howdy all!

Does anyone know how you would go about unlinking channels of the PRR Varimu??

There's seems to be alot about how to optimize the stereo controls but not much chatter about the way to get the unit to function as two independent channels.

Any clues?

Thanks
Illumination