Clamping diodes to limit crossover distortion?

[Chuck H]

Either the forum or my computer circumstances aren't offering "like"s right now. Just want to say that Juan gets one from me.

The diodes seem to be there for the reasons stated above. As to the voltage spec., possible typo? My system, which isn't actually my mod but a small addition to and a formula for using the Paul Ruby mod, actually clamps the negative voltage swing immediately after the tube is driven into cutoff so the coupling caps can discharge when no useful signal is happening. It works a treat and some amps that use el84's benefit from it because they can sound so buzzy and fizzy compared to other tube types under the same overdrive levels. I don't use it for all amps and never with big bottles for the reasons Juan stated. Unless you have a fizzy el84 amp there's really no need to implement the "chuck h"/Paul Ruby mod. I think you would do better to reduce screen sag by any means (smaller filter, diode drop instead of resistor, branched PS instead of a series rail resistor, etc.) and heat the bias a bit using tubes that can handle it. That's where I would start. Not mentioned yet, but I don't think that amp is exactly revered for it's OD tone. It's more of a loud and cleaner sort of amp I think. Trying to make it into something it's not is like racing with a dump truck. It has four tires like a race car, but it's not a race car.

[nsubulysses]

You guys are great but unfortunately I still don't understand exactly what the diodes from control grid to ground do, only for the fact that I see no difference on the scope screen and I hear no audible difference with 1N4007s clipped in. Is keeping positive voltage off the control grid good because it limits bias excursion or because if signal level exceeds negative bias supply level it causes the PI coupling caps to charge and causes blocking distortion?? Or something else? SOrry I know it's been explained a couple times but I don't totally understand. I am referring to the diodes 1N5062 in the model T schematic in post 1.

Anyway, this is basically what i'm working with.




Chuck, your mod has made you legendary. now you will have to explain it over and over to people who don't quite understand it. Maybe this amp isn't a good candidate and it's a look inthe wrong direction anyway. ALso, you are thinking of the 1970s model T which is loud and clean with KT88s and ultralinear output. I'm thinking of 1990s model T made by Fender which is just a modern high gain amp with channel switching and clean and gain channels, 100W with 6L6s

This amp sounds good. Just trying to perfect it.

Power tubes are 6CA7 so .026 x 586V = 15.23W / 25 = 61% plate dissipation (sorry juan)

[Enzo]

You see no difference with and without? OK, in your tests, are you driving the tubes with enough signal level to drive the grids into conduction? As protection devices, they only do anything when conditions exist that require the protection.

Like seatbelts in your car, normally trhere is no difference wearing or not wearing them. it is only when you run into something that they determine if you explore the face/windshield interface.

[nsubulysses]

I will have to test when I have the amp back open again but I guess probably not. Just guessing though. Perhaps my -51V bias voltage is a lot to overcome, if I'm thinking about it right. Would these diodes be more useful in a circuit that only has a -30v bias voltage ?

[Enzo]

How many amp circuits have you seen? How many had this particular diode thingie? They saw fit to add them, but the vast majority of amps don't have them. Don't obsess over it. I don;t see anything else special about this amp, amps all tend to be very similar under the skin.