I always find Gnobuddy's input informative, useful, and accurate. His analysis and recommendations make good sense to me! My own experiences with VVR have been on boards, not tag strips, and even then, I've had to adjust layout on a few builds to make them work perfectly.
It will be interesting to hear about your results, razorrick1293.
It will be interesting to hear about your results, razorrick1293.
Thanks for the kind words, GTK!
These dang newfangled power MOSFETs are super-fast, capable of switching at hundreds of kHz, and that can cause major headaches for us DIY types!
A VVR has no reason to be super-fast, so I wonder if it would behave better if one slapped a big fat capacitor (say 1 uF) between source and gate, right at the MOSFET terminals? With the 100k gate resistor shown in post #1, the RC time constant is 0.1 second, so that might slow the MOSFET down to a sensible speed for this particular application.
Incidentally, a 7W incandescent night-light bulb (I still see them at Home Depot and in dollar stores) draws a bit under 60 mA at 120V. Two of those in series might make a useful dummy load for the VVR, and you can dial B+ up to 230V or so without damaging the bulbs.
If you need to go above 230 V, three of those 7W bulbs in series should do the trick. That should be good for up to 360V.
-Gnobuddy
These dang newfangled power MOSFETs are super-fast, capable of switching at hundreds of kHz, and that can cause major headaches for us DIY types!
A VVR has no reason to be super-fast, so I wonder if it would behave better if one slapped a big fat capacitor (say 1 uF) between source and gate, right at the MOSFET terminals? With the 100k gate resistor shown in post #1, the RC time constant is 0.1 second, so that might slow the MOSFET down to a sensible speed for this particular application.
Incidentally, a 7W incandescent night-light bulb (I still see them at Home Depot and in dollar stores) draws a bit under 60 mA at 120V. Two of those in series might make a useful dummy load for the VVR, and you can dial B+ up to 230V or so without damaging the bulbs.
If you need to go above 230 V, three of those 7W bulbs in series should do the trick. That should be good for up to 360V.
-Gnobuddy
You need to measure the voltage on the mosfet gate.
if its low there it will be low at the source output.
The zener diode turned out to be shorted. Vgs = 0 volts, no wonder the MOSFET wouldn't turn on!
-Gnobuddy
Now that is interesting!
I had been assuming the gate would fail open-circuit, or short to the source. But what you just said suggests perhaps a drain-to-gate short occurs when the MOSFET fails. (That would put the zener diode in series with the entire MOSFET current, causing it to fail.)
-Gnobuddy
A good grounding scheme is necessary for a stable amplifier. Have you read Merlin Blencowe's write-up on grounding? Although it's a chapter from his tube preamp book, he was kind enough to make it free to download: The Valve Wizard
IMO the book (Designing Valve Preamps for Guitar and Bass, Second Edition) is well worth buying, if learning more about tube guitar amps is something you're interested in.
-Gnobuddy
I've just bought that book (waiting for it to arrive) and yes I have read that section and I've followed it as closely as possible. What I don't understand is, the non ideal grounding worked better. I can only think of 2 ground wires running parallel where before there weren't. And I don't understand why it would oscillate at medium power and not at high
I started out building circuits with germanium transistors when I was about 8 years old. Those transistors were slow enough that I never had oscillation problems.
A few years later came silicon transistors, but with low input impedances and limited voltage gain, I still had no oscillation troubles.
After that came integrated circuits. My first was the 741 op-amp, also slow, also stable and easy to use.
A few more years went by (I think I was 18 or 19 by then), and I tried to build a (cassette) tape preamp with an LM381 chip. This time it was a different story! The LM381 has huge voltage gain (112 dB), it's fast (15 MHz unity-gain bandwidth), and there are two channels to cross-couple and create even more havoc. And it's tiny - so there's lots of gain in a small space, practically a recipe for unwanted oscillations via stray capacitance.
So my preamp oscillated like crazy, and I fought that problem for some considerable time before I finally beat it. After trying all sorts of ideas, I ended up with a grounded metal screen (cut from a tin can!) positioned flat over the top of the chip, wide enough to cover the entire chip as well as the leads emerging from its sides. That finally did it, and the preamp went into a DIY cassette deck I was building.
What I'm getting at is this: there is a learning curve to coming up with a good layout. It's partly science, partly art, and some requirements conflict. When things go wrong and the layout isn't right, there is rarely an easy fix. The instability will manifest itself in many different ways, and may stick around stubbornly while you try one possible solution after another.
Most inexperienced DIY guitar amp builders side-step the problem altogether by sticking closely to a tried-and-tested layout they found online. Those who don't do that, usually have a long learning curve ahead of them, as I did.
-Gnobuddy
I think you'll be very satisfied with it. I myself think very highly of that book.
I've been designing and building solid-state audio circuitry since I was a child, but tubes were entirely new to me, and the way they're used in guitar amps is quite weird, completely unlike normal audio electronics design. Blencowe's book was an enormous help when I was trying to get my head wrapped around the bizarrely weird world of tube instrument amplifier design.
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.