VVR question

[daz]

I just bought the parts needed for the schematic below today. Darn mosfet was $12 ! Anyways, i was going to impliment it just as shown, but i wanted to scale the PA only. However, not having a schematic of how to do that i decided to do the whole amp as the schematic shows. But the more i think about it the more i feel thats not going to work near as well because of my high gain/low voltage preamp.

so heres the question. How would i go about using the circuit below to scale the PA only? I cannot see how i could do that without adding a second main cap can and a second choke. Not that i would even consider doing that, nor could i due to lack of space. But let me explain what i mean and if you can see the error in my thinking please enlighten me. My amp is configured pretty much like a JCM 800 but it's cathode biased. The first thing the B+ sees is the OT, then the 32/32uf cap can with a choke between the cans. after that it goes to the screens via 1k resistors. heres where i see the problem....if i were to scale the PA, the choke and can would be included. In other words, the B+ would go onto the scaling circuit before the OT, then the scaled B+ would come out before the OT. Now if i seperate the PI and preamp from the scaled B+ i would have to take it from before the can and choke at the point where the B+ goes to the scaling circuit. I would have to "Y" the B+ at that point, one side going to the scaling circuit and the other to the PI. Of course i would use diodes to isolate them. But the point here is that the PI and pre are now seeing the B+ as before, but WITHOUT the choke and cap can before them.This is what i don't understand. how can you do this without adding a second can and choke for the pre and PI?

[Enzo]

What was the original FET the NTE cross came from? NTE parts typically cost 5-10 times more than the real parts they replace.

[daz]

I don't know the generic number. I only saw the NTE#.

[Bruce / Mission Amps]

What was the original FET the NTE cross came from? NTE parts typically cost 5-10 times more than the real parts they replace.

I've never really split off the B+ rail but here is a list of power MOSFETs that work:
STF9NK60ZD, FQA8N80, FPF6N80, FQI4N80, 2SK3675, NTE2377, MTP6N60E, IRF740B and ....I've had good luck using very inexpensive IRF830s in smaller, low voltage, lower powered (<20w) amps... the trick is to use a power MOSFET that can handle like 500v to 800v and at least 75 to 200 watts, depending on how big the amp is.

[daz]

Well, i used the NTE2377 and it buzzes so bad after about 1/10 turn of the VVR knob it's unusable. Dana said it could be the mosfet because he's found that the 2377's sometimes do that and switching to a NTE2973 fixed it. In any case i'm not going to keep buying these things and wasting my money unless i can be sure it IS the 2377 thats to blame. i will look thru my parts and see if i have one of those on the list tho....thanks Bruce.

[Bruce / Mission Amps]

Well, i used the NTE2377 and it buzzes so bad after about 1/10 turn of the VVR knob it's unusable. Dana said it could be the mosfet because he's found that the 2377's sometimes do that and switching to a NTE2973 fixed it. In any case i'm not going to keep buying these things and wasting my money unless i can be sure it IS the 2377 thats to blame. i will look thru my parts and see if i have one of those on the list tho....thanks Bruce.

Well, trust in what Dana says 'cause he is the one that has done most of the legwork, turned me on to all this and supported many of us in the techno discovery part of the whole thing.

[daz]

I do, but i re-read his post and realized i mis-read it in the first place. He said it was a 2sk3675 that buzzed and he replaced it with one of the 2 he uses. Not the one i have tho, which is the other one he uses. So who knows. I sure wish i could try it because what i CAN hear behind the lous buzz sounds like it may be good. But it's hard to tell behind all that noise. Really bums me out because i was looking forward to this for a long time.

[Steve Conner]

Did someone say "0.1uF capacitor from the MOSFET gate to ground"?

The FET may need a 100 ohm grid, err, gate stopper to prevent parasitic oscillations with the cap installed, ie, don't connect the cap straight to the gate pin: put the 100 ohm resistor in between the gate and everything else.

But it should kill the hum deader than Monty Python's parrot, no matter what MOSFET part you use.

With the cap, the circuit may have enough of a smoothing effect that you won't need the extra choke and capacitor for your screen supply: just connect the screen supply straight to the output, and leave the existing choke and cap to smooth the preamp. Note that the circuit will lose its smoothing effect when the voltage control is dimed, so you'll need to switch the screens back where they were at full power.

Or arrange a separate RC smoothed supply for the preamp and leave the choke/capacitor on the scaled rail.

[HTH]

The FET may need a 100 ohm grid, err, gate stopper to prevent parasitic oscillations with the cap installed, ie, don't connect the cap straight to the gate pin: put the 100 ohm resistor in between the gate and everything else.

I'm sure RG suggests the same thing in his Mosfet follies section of Geofex when using a Mosfet as a CF.

The VVR circuit looks very promising, I have some IRF740s here to try once I get my arse in gear. Think they cost me under a quid each from RS.

[daz]

Actually guys, i found the problem, tho it makes little sense. I found it by searching for VVR threads at 18watt.com. Someone was having the same buzzing issue and said if they loosened the screw on the mosfet it stopped. Sure enough i tried it and the buzzing stopped. Why, i have no idea. I checked the conductivity between the chassis and the metal backing on the mosfet and it was completely open, so it wasn't somehow shorting thru the insulator. so i'm stumped. however....ask me if i care ! I really wasn't happy at all with VVR once i got it working. i tried scaling the whole amp and the preamp gain went from the high gain pre i designed to almost totally clean at low VVR settings.I guess i should have expected that, but i never thought about it, and i've never heard anyone say that. i guess most people use these in smaller amps that get thier grind from turning them up.
Then i scaled the PA only. That didn't work either tho for a different reason. The master volume at 9:00 is still a bit too loud for home practice for me. But unfortunately with the VVR all the way down the volume at 9:00 isn't affected at all. It when i turn it up more that you see the affect. With the VVR at zero the amp's volume barely changes from 9:00 to full up. But that does me no good for my purpose.

So yeah, it works now. But it's not for me. There are other solutions that work far better for my purposes, at least on this amp.

[Bruce / Mission Amps]

I do, but i re-read his post and realized i mis-read it in the first place. He said it was a 2sk3675 that buzzed and he replaced it with one of the 2 he uses. Not the one i have tho, which is the other one he uses. So who knows. I sure wish i could try it because what i CAN hear behind the lous buzz sounds like it may be good. But it's hard to tell behind all that noise. Really bums me out because i was looking forward to this for a long time.

I just looked and the NTE2973 devices are the ones Dana gave me to mess around.
Did you build the circuit that also decreases bias voltage to keep the power tubes linear at that new B+?
Also, I never did hear any buzzing....
To be 100% honest, I can see where this circuit could be attractive to some but I'm not knocked out but the whole thing anyhow.
I prefer a more natural approach with smaller amps and smaller power levels, over driven to get that sound... I guess what I'm saying is that a true craftsman has quite a few tools in the tool box... not just one or two that ultimately, do everything equally mundane.

[daz]

Did you build the circuit that also decreases bias voltage to keep the power tubes linear at that new B+?

No, because the amp is cathode biased. as i'm sure you know, the one that tracks bias is for fixed.

To be 100% honest, I can see where this circuit could be attractive to some but I'm not knocked out but the whole thing anyhow.

same here, but i was surprised that i wasn't because all i ever hear are raves. I have a jack in the amp thats nothing more than a shielded cable that goes to the wiper of the master. i use it to give me a tad less volume by putting a cable from it to my little rack that houses my tuner and a couple other necessities. And the cable goes to a pot that is simply wired as a simple resistor to ground and is switched in or out via a relay. I set it so the volume is a tad quieter when it's switched in, then use that setting normally and the switched out setting for slightly louder solos and such. the reason i'm mentioning this is that for home practice it's even more effective than VVR for getting a decent sound at super low volumes. Not perfect of course. If it was i wouldn't have been so anxious to try VVR. But it works better for my purposes, so goodbye VVR. I would never in a million years thought it would work better than VVR with all the raving that goes on about it.

[Steve Conner]

Well, the circuit posted only cuts the B+ in half when the pot is turned down as far as it'll go. That cuts the output power to about a quarter.

So it's not surprising that it doesn't have much of a subjective effect. Remember that you need to cut the power by a factor of 10 to make something subjectively half as loud.

I lately experimented with an amp that has a 3 watt SE output stage, and a "master volume" that's a big 3-position switch. It selects the output stage alone for 3w, a 10dB attenuator between OT and speaker for 0.3w, or a +10dB booster (an extra solid-state output stage similar to a Thomas Vox, driven by the existing tube one) to get it up to 30w.

And indeed each position seems about "twice as loud" as the last one.

[daz]

Thats Dana's exact circuit tho, and what everyone is using. In fact, there are some that have changed the R values to make it attenuate less because they say it goes too low. Thing is, it's the PA scaling it doesn't lower enough. If i scale the whole amp, turning it down all the way almost shuts it off completely. That may give you a clue as to why when i scale the PA only it doesn't go low enough, and if so i'd be interested to hear and might give it another try if that can be rectified.

[HTH]

I lately experimented with an amp that has a 3 watt SE output stage, and a "master volume" that's a big 3-position switch. It selects the output stage alone for 3w, a 10dB attenuator between OT and speaker for 0.3w, or a +10dB booster (an extra solid-state output stage similar to a Thomas Vox, driven by the existing tube one) to get it up to 30w.

I'd love to see a schematic of that - sounds interesting.