Tube amp knowledge for dummies.

[misterhinkydink]

How about you post your explanation instead of correcting Kindess' post? That might be easier and I'd love to read it honestly. Dan

 

 

Kindness's explanation is based up a tube circuit being constant current and that's not how tubes work. Varying the current through the output transformer is what causes the speaker cone to move. It's that simple. I know he knows this because he doesn't make the same claim in the Marshall write-up.

 

I think the question was what would happen with a mismatched impedances on the output. Nothing if the speaker impedance is the same or smaller than the amp's nominal output impedance. If the speaker impedance is greater, output transformer design maximum voltages can be exceeded which can cause the output transformer to fail. So lower impedance is OK, even a shorted output won't hurt anything, but higher is bad.

[Kindness]

Kindness's explanation is based up a tube circuit being constant current and that's not how tubes work. Varying the current through the output transformer is what causes the speaker cone to move. It's that simple. I know he knows this because he doesn't make the same claim in the Marshall write-up.

 

 

I wrote what's in the JTM45 write-up. The post above was something I wrote previously based on direct input from an EE/amp designer off this board (and copied and pasted into this thread). I still don't feel like I know enough to tell him he's full of {censored}. I'm still just learning.

[L. Ron Hoover]

Correct me if I'm wrong but the transformer primary acts as the plate load for the power tubes so changing the load impedance changes the load on the plates under AC conditions which should change the output characteristics of the amp. I don't see how that would change the voltage seen at the plates, though as that's a function of the supply voltage and whatever drop there is across the DC resistance of the transformer primary.

I think that you would model a tube output stage without a transformer as a constant current source in that the high impedance output would be susceptible to Ohm's law interaction with the load that would cause a larger swing in output voltage versus output current into a variable load impedance. For example, if you had a 100 Ohm output impedance and a 4 and 8 Ohm load, the voltage at the top of the 4 Ohm load would be about 50% what it would be at the top of the 8Ohm one, but the total current would only change by 4%. That's part of the transformer's job, to convert that quasi-current source into a quasi-voltage one with respect to typical speaker load impedances. That's not the same as thinking of a tube as a constant current device, as like any device the output current is always proportional to the output voltage.


As far as I can figure the biggest issue with connecting higher than rated impedance loads to tube amps, culminating in the ultimate sin of no load, is inductive flyback in the transformer, which can be quite severe with infinite load impedance.

[Narcosynthesis]

So I was making a private exchange with Kindness because as some of you know I have a vintage Ampeg V4B coming my way within the next couple weeks. This was an exchange where at the end I was honestly more confused that when I started; which is saying something because I know very little about tubes.. Instead of going back and forth I figured I'd post a thread because there might be other people that want to learn. My question is about biasing and tubes. So this is what I thought biasing was: Basically setting the amount of current running through the tubes based on the type of tubes and their tolerances. I thought most tube amps had to be biased when they were re-tubed if not more often. I know some tubes had a fixed bias (400+) but apparently there is more to it. I thought that tubes are matched in pairs. So, feel free to post your own tube questions as well. If you can explain this to me as well I'd be really grateful. Also any advice on general upkeep that I can do myself would be nice. Thanks! Dan

 

 

To try and give my explanation of the original question...

 

A valve as a fairly old technology varies a lot more in tolerance compared to something like a transistor. Biasing an amp is basically adjusting the amp to use the valves at their most efficient - an amp biased 'cold' means the valves are not being used to their fullest and so you are never reaching the point where they sound their best. On the other hand biasing the amp 'hot' means the valves are being run beyond their intended settings, so you will often get more gain and distortion as you are running the valves closer to the point of distortion (at lower volumes, so you don't need to crank it quite so much for the same sound) with the downside that you are putting more stress on the valves and wearing them out quicker.

 

Valves are matched for the same tolerance reason - in the amps we are looking at one valve amplified the positive side of the signal wave, another valve amplified the negative half, so you get a matched pair to make sure both valves are within tolerance together and will sound the best, if the valves are too far out you can have problems where the two halves of the wave interact (instead of having a smooth since wave, you get a stepped one) which can lead to distortion.

 

Many modern amps don't give you the option to bias the valves - I am mostly guessing here but part of it I would assume to be better modern manufacturing meaning an amp can be setup for a particular valve, and you should be able to buy a new set of valves within tolerance, older valves would vary more and so need biasing to stick within a certain tolerance band.

 

As for maintenance, you shouldn't be inside an amp unless you know exactly what you are doing - the caps can easily hold 400v+ even after the amp has been off for a long period of time (dependant on amp model and design, but I would most definitely play it safe).

 

In normal use, make sure to use appropriate cabs for your amp, and give the valves a few minutes to warm up before hitting them hard - A lot of amps give you a standby option which lets some current to the valves to let them warm up without hitting them with the full signal - hitting them with the full current at the start before they are warmed up is a good way to shorten the lifespan of a valve. If you don't have a standby mode, perhaps turn it on but with the master at 0 and no input plugged in so as not to stress the amp straight away.

 

For those gigging or moving the amp to practise and so on, try and let the amp cool down before moving it too - when a valve is hot it is at its most fragile, so thumping it down a set of stairs isn't going to be the best idea. Try and let the amp cool down a bit before moving it if possible (which it isn't always when you are trying to shift stage sets and so on, at least be careful not to knock it if you can)