Modified V76 is Kicking My Butt

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
G

Guest

Guest
I have been trying to build a modified version of the V76, thinking it would be a good thing for my son to use in his radio production studio. Well, it is kicking me butt badly.

First, the circuit has so much gain that it becomes easily unstable and oscillates in both low frequency (about 2-3Hz) and high frequency (about 200KHz). Grid stoppers helped the LF oscillstion and I can play with small caps across the feedback sections to stabalize the HF oscillation but it is still way too touchy.

I lost most of my hearing so I rely on oscilloscope readings for the general picture of how the circuit is working. I rely on my friend Ron, who runs a small recording studio, to actually hear what it sounds like. Well, after wrestling with oscillation for a while, I took it over to his studio for preliminary checks, hooked up a mike and found "highs rolled off, kind of midrangy with an emphisis on lows".

It appears the feedback gain circuit between the first and second stages is the main culprit in rolling off the high end. I am puzzled as to how it is doing this and have come here for some ideas. Also, as I'm sure you are aware, this circuit has so much gain that it ia nearly unusable without a really quiet stidio.

Here is a schematic of what I am trying. And and all help is appreciated.
http://users.adelphia.net/~thomasholley/Microphone/Modified%20V76.gif
 
Thomas,

In your drawing, on the second block (EF40/EF83), you show the feedback connection going to the input grid of U3. I believe this should instead go to the suppressor grid/cathode connection. Otherwise, I believe you've created a positive feedback condition.


OldHouseScott
 
Very good Gentlemen!
Here is the correct hookup:
v76_trouble_1.jpg


That will most definately make an improvement, if indeed, you had it mis-wired.
That will cure the high end oscilation.

To cure the motorboating, lower your plate voltage and drop the screen current thru the second stage.
This is where your motorboating is originating from.
The Lundahl choke probably has less DCR than the original Tele 16 million turn jobby.

PS:
That 80 k resistor needs to be adjusted for minimum harmonic distortion.

cj
 
Thanks a lot for looking at my circuit. I had it misdrawn and not miswired, thank god. I was actually using the real V76 schematic to go by and drew up my changes. I basically took out the tone shaping circuits between the first pair and second pair of tubes and left the rest alone.

I added individual decoupling for each stage because I was concerned about noise.

CJ, the Lundahl chokes have about 2.4k ohms DCR. How does this compair to yours or the original ones?

Jakob, the 80k on the input of the second pair is there only because the original schematics had iether an 80k or two 40k resistoed in the tone shaping section so I just stuck that in for no other reason.

Thanks again for the help.
 
The original chokes had quite a bit more dcr, you can figure it out with ohms law, i think the original schematic has the current and voltage thru the choke, so just divide the voltage drop across the choke by the current and you have your ohms.

Post voltage readings for that second stage so we can compare it to the original.

Sometimes, you want two stages connected to the same decoupling cap, it can help cancel oscillation as the ripple cancels.
I don't know if that applies to you, but when you get stuck, you want to have as many options as possible for straightening out a really tough problem, and believe me, motorboating can be one tough sob to stop.
I spent a whole month getting my V76 so it wouldn't motorboat at even the highest gain setting with a condenser mic. I must have changed every component in the amp at least twice! Not to mention the altering of the mechanical layout, especially the feed and output caps. Your transformers are obviously different than the originals, so you might have to make other components that wotk with the iron non stock also.
cj
 
CJ, I am not thinking. I have the schematic with current and voltage so DCR is easy. My copy is a bit blurred but if the current on the first choke is 0.6mA, that maked DCR about 112K. Pretty high. The E83F choke is about 9.3k DCR. More reasonable.

What worries me most is what my friend said about highs being rolled off in the first pair. The input is good out of the transformer to the grid of the first tube, but the signal is rolled off out of the second tube. The feedback gain circuit must be the culprit but I am not sure how it is doing this.
 
You want the high end roll off.
That's part of the genius of the V76.
The tone is compensated as the gain is changed.
It took me a year to figure out how, thats how slow I am.

I am not going to tell you how. That would spoil the fun!
Hint: look at capacitors.
 
So, I want the roll off? It will take me a long while to figure that one out. I have two questions then.

1. Did you use the schematic values for gain feedback? I have not tried to mess with these.

2. Did you have any of the tone shaping stuff between the sections?
 
I ditched the pi filter, thats the pf caps and little inductor.
Funny story, I had Ollie send me that inductor just so I would know if I were throwing out part of the secret sound of the V76.
When I recieved the inductor, (he probably throws out ten of those a month when he does repair work), I measired the inductance. It was way off. The glue that held the ferrite pot core togeather had failed. Understandable since the thing was probably 40 or 50 years old, and it gets Hot inside the V76!
The inductance was very low, so it really wasn't doing anything anyway. I wonder how many V76's out there have the same problem. It may actually be a blessing, as that filter rolls off the top end.

Anyway, long story short, I ditched the pi filter, bit I do have two tone switches, one bass boost and one high end cut. They are fairly subtle, no radical sound changes when they are switched in and out.

The roll off during different gain settings gives the amp that velvet , silky smooth sound . I used the exact same resistors in the gain circuit.
The tone switches are just a cap and resistor to ground , and a different value for the coupling cap from stage 3 to stage 4.


cj
 
Thanks loads. I am going back to work on this and will measure some voltages and currents and report back.
 
I updated my schematic. To try to reduce low frequency oscillation I added 10k grid stoppers to stage 1, 2 and 4. I added some larger decoupling resistors to lower plate/screen voltage and documented the measured voltage.

For current, I get about 0.5mA on stage 1 and 3, about 3mA on stage 2 and about 12mA on stage 4.

With no input or output connected, I get LF and HF oscillations on several gain settings.

I'll go back into it tomorrow.
 
The first thing I do if encountering oscilation is get the amp to oscilate, then use chopsticks to move wires around. You can sometimes isolate the feedback to a certain area if you see a change on the scope while doing this.

I had an electronics teacher in Highschool who could fix things like this in a hurry.
He had 30 kids to help, so he would rush around and just start moving grid and plate wires around til it stopped.
Man was he fast. And good!

cj.
 
CJ, your suggestions forlowering screen current and adding series resistance to the chokes worked to tame the LF oscillations. Thanks so much. I was getting a little frustrated and needed a push to keep from sledge hammering the works.

I still have a HF oscillation at about 0.5V a aroundt 800kHz. I will tackle this in the morning.
 
I played around today to find the key to turn off the HF oscillation. First. I added a 10p cap across the first gain feedback resistor which took care of the oscillation on the input side.

Next I used a box I made with a bunch of pots in series to take the place of a decade box to set the feedback resistance for the output pair. After I got everything set I measured the resistance in my box (137k) and put 2 series connected 68k in for the feedback resistor, turned it on and...still oscillates! Well after much head scratching and retesting, I measured the capacitance in my decade box and wire clips and had about 40p. I testes values of caps and selected 22p as the best result to parallel with the feedback resistor.

I ran square wave and sine wave signals through the V76 at all gain settings with no oscillation. I compared the input right out of the LL7903 input transformer and the output of the LL1680 output transformer and signals matched pretty closely. Now if I could hear worth crap I would do some listening tests. I'll report when I can get the amp into the studio.
 
[quote author="Winston O'Boogie"]...I'm just curious as I'm planning on building a couple of these myself now that I just received some appropriate iron for the job.[/quote]I'd be definately interested in building one of these as well, maybe a META-V76 thread is in order.
Kelley
 
I´m also planning on building a V76, but it´s been a pain finding the right chokes (I really cannot afford TAB-Funkenwerk chokes).

I´ve contacted Haufe some time ago and they said they could produce the exact chokes for the V72 for 106EURO a piece (30 minimum order), but not the V76 ones. I was planning a group order but there was no interest at RO forum.

So, if the Lundahl chokes will works fine, I would definetly try it.
 
Back
Top