Q re: tube rectifiers and sag

[Leonc]

Is there a particular rating (e.g., maximum plate voltage) that will be indicative of how much an amp is likely to sag, given a specific set of preamp and output tubes?

E.g., if you have two similar amps--e.g., both powered by a 6V6 and both are using 12AX7s--but they use different rectifiers...can you state in general that the one with the rectifier in which _____ is rated (higher/lower) is going to sag more?

Or do you really need to know more about the rest of the circuitry (e.g., are both triodes being used by the 12AX7s) before you can make this kind of generalization?

[donnyjaguar]

Have a look at the conduction curves of GZ34 and 5U4G to get an idea of what kind of "sag" you'll get. To complicate matters the design of the collective power supply will also play a roll in this. The rest of your missive doesn't make sense to me so perhaps try wording it differently.

DJ

[Leonc]

dj - thanks for trying to interpret my layman's twisted and partial (make that very partial) understanding of tube amps.

What I'm asking is...is it possible to predict how much an amp is going to sag given a specific type of rectifier tube (and holding other things--like the other tubes being used--constant)? If so...how do you identify that kind of rectifier tube?

[fullerplast]

Generally, higher internal resistance of the rectifier and/or lower value filter caps = more sag. The GZ34 will sag less than a 5U4 in the same circuit for that reason.

[Leonc]

fp - thank you. That's the kind of info I was looking for. So it will help to know the value of the filter caps. And if you do, then looking at the internal resistance of the rectifier tube.

So...let's say I've got the RCA Receiving Tube Manual...which tube rating tells me what it's internal resistance is? In my ancient manual, they include various ratings for many of the tubes (but not all)? E.g., could it be Effective Plate Supply Impedance per Plate? That appears to be the only rating that is measured in ohms for rectifier tubes...

[Old Tele man]

re: "Have a look at the conduction curves of GZ34 and 5U4G..."

...or, you can mathematically 'approximate' the rectifier plate "voltage drop/sag" using the Child-Langmuir 3/2's Law for diodes:

Ip = G * Vp^(3/2)

...and backsolving for Vp:

Vp = ( Ip / G )^(2/3)

...where:
Ip = Plate current, amps
G = Tube Perveance, amps-per-volt^(2/3); ~ 0.00321 for 5AR4/GZ34
Vp = Plate potential, volts ("drop/sag"); ~ 17V @ 225mA for 5AR4/GZ34

...so, if the total amp current is 200mA at full power, then you can estimate the 5AR4/GZ34 voltage drop/sag to be about 16 volts:

Vp = (Ip/G)^(2/3) = (0.200A/0.00321)^(2/3) = 15.7 ~ 16Vd.

[fullerplast]

fp - thank you. That's the kind of info I was looking for. So it will help to know the value of the filter caps. And if you do, then looking at the internal resistance of the rectifier tube.

You can basically look at the plate voltage drops between rectifiers to establish an order of internal resistance (and sag). The order of magnitude is usually in hundreds of ohms. For most of the amps we use it will look like this in order of least resistance to most:

GZ34
5U4
5V4
5Y3

You can fine tune sag even without changing rectifiers by adding a series power resistor, say 100ohm, in the circuit to simulate a change in internal resistance. Even easier, if you want to experiment in this area, is to get a few Weber custom copper caps with various series resistors built in. Beyond that, you can try decreasing or increasing your filter cap values to increase/decrease sag. Both these factors really affect the "feel" of an amp at higher volumes and give an amp some natural compression (at the expense of tight bass and perhaps ghost notes and hum if you go too far).

[Leonc]

Interesting. Where I'm coming from is that every now and then, I stumble across an oddball old amp. I love the tones of some of them produce...but I don't love too much sag and some of the oldies get really saggy. I little bit is okay, but I can't hack really squishy feeling amps. Let the harp guys have 'em :). So I'm trying to get a handle on if there is a way to determine how much the thing is going to sag before even trying it. It's also nice to know that it can be diddled with...

[fullerplast]

Don't like the saggy old amps huh?;) Look for big power trannies, solid state or GZ34 rectifiers, and good sized filter caps. Traynors YBA-1's for example.

5Y3's, with 16uF caps and PR or DR sized PT's are gonna be saggy...and that can be a very good thing for some applications!

[billyguitar]

Change the amps to cathode bias to enjoy the most sag. I've done it on some Fenders and depending on what rectifier tube you use, you can dial it in pretty good.