How to properly bias an express......

[redshark]

Ok this is interesting here. In the old Trainwreck pages Ken mentioned to bias the express amp using the negative voltage method setting it to -30vdc.
If we see the Trudy schematic the resistor before the bias pot is a 56k and the notes on the schematic say adjust to -30vdc. This amp was made in 1985.
In the Kelly(made in 1990) schematic the resistor before the bias pot is 47k and the note says -33vdc.
Glen Kuykendall once mentioned that ken told him to bias -32vdc.
Here the normal practice in the forum seems to be doing it by current draw and adjusting to 45 to 48mA.

What is the correct way????

In my amp with a bias resistor like Kelly 47k a bias setting of 45mA gives me -31.5vdc....
Has anybody checked this thing before?

[paulster]

I've found that about -30V to -32V gives you a bias figure of 70% on regular plate voltages and NOS Siemens EL34s (when you factor in the static screens dissipation figure).

I go for 70% on the plate measured through 1ohm resistors, and the Express definitely favours this kind of hot bias.

[redshark]

Paulster please explain me how the screens get involved here....

[paulster]

If you add 1ohm resistors at the cathode connections to measure bias across (as Komet et. al. do when they add bias test points) you'll be measuring the cathode current, which includes the plate current, screen current and grid current (0).

So your reading will be slightly higher than the true plate current because of the small amount of screen current in there as well.

If you measure the voltage drop across the screen resistors you can calculate the screen current and subtract this from the cathode current figure to give you a true plate current.

Then get that up to 70% of maximum dissipation and you should be set.

[redshark]

I didn't add a 1ohm resistor, I use bias probes.

So my plate voltage is 408 and screens 388. How does that work?

[paulster]

A bias probe includes the 1ohm resistor, so you're measuring the cathode current.

What you want to do is measure your screen voltage supply (pin 6 on one of the output tubes) and then the actual screen voltage on each of the tubes (pin 4). Be very careful if you're not used to measuring voltages inside the amp and especially on the tube sockets.

You can then work out the voltage dropped across each of the two screen resistors and therefore the screen current (I = V / R).

Subtract this from what your bias probe is telling you and you'll have the true plate current.

[redshark]

Ok, I will give it a try. Never heard about this. I always used plate voltage and current to set bias. Thanks!!

[redshark]

Ok Paulster I took voltages...

pin3 408vdc
pin4 385vdc
pin 5 -31.7vdc
pin6 391 vdc

bias current in the probe now set to 45mA

how does this work now? sorry....

[paulster]

So your screens supply is 391V (that's common to both tubes),

On the tube you measured voltages on, your screen voltage is 385V, so that gives you a screen resistor voltage drop of 6V (391V - 385V).

Assuming your screen resistor measures exactly 1K (measure them both with the power off and the capacitors safely drained) then that particular tube's screen has 6mA flowing through it (6V / 1000ohms = 0.006A, multiplied by 1000 to get a figure in mA = 6mA).

The screen current will vary a little between tubes so you want to measure them both, but this demonstrates how to get the figure.

On this particular tube, with the bias probe showing 45mA you actually have 39mA of plate current. This means that although the bias probe is telling you that you are at 73% dissipation, you are actually at 64%.

That screen current figure looks a little high though, so if you have clip-on probes for your meter I'd recommend connecting them either side of the screen resistor and measuring the exact voltage drop rather than the voltage either side referenced to ground.

[redshark]

Thanks!!! I got it now, I will do the same with the other tube.
Love this forum!! always a chance to learn something new.

[redshark]

I'm surprised nobody else commented on this thread.
Anyway still seems to me the Trainwreck world always has this misterious vibe. Like things are never fully open for everybody. I know there's people here that used to own real express amps but the only guy that fully discusses his amp has been Glen, I love the tech video of his express. Still he has never showed pics of his amp. The only pics out there are Francesca which has a couple of "errors" and the "reject" which wasn't performing properly. Then once we build our clones all of them seem to have little issues, mine are oscillations and not in "performing settings" which is good. Still each time I move any wires around the pots or in the V1 and V2 the amp sounds diferent, sometimes, darker, sometimes gainier, sometimes microphonic rattles with certain notes, then move wires and the rattle is gone. Very temperamental build.

[Colossal]

Hi 'shark,

It's my belief that the Zen of the Express is in the journey, not the destination. Part of doing an Express build is the research. As far as things not being completely transparent, I dunno dude, I think all is revealed here (it just takes a bit of digging and many hours of reading and careful notes). The circuit is quite simple (as Glen discusses in his amp tech talk video) yet confounding complex to get right. Glen and many others here have been very forthcoming about the inner workings, generous with details and analyses, and as you noted, Glen has provided volumes about his personal Express as well as his experiences in building his clones. I believe he has withheld photos of Orphan Annie out of respect for Ken.

Just some thoughts...

Regards,
Dave

[M Fowler]

Red,

I think everything has been very open about the TW amps and I have spent many hours starting from the first posts 2006 or somewhere back there for reference. No secrets about the TWs I can see. Ron Worley spent an enormous amount of time on the Express build guide. Ron Worley and others also spent hours on layouts, schematics, and BOMs.

The Express takes some time to properly lead dress, parts placement, using the specified components and much more.

Besides it was a great learning tool for you right?

Mark

[redshark]

Yes Mark you are right, and so are you Colossal, I think that the amount of information here is plenty and everybody is always willing to share but still, we haven't heard all the express amps, we haven't seen all amps gutshots we haven't tested all of them I'm talking about real ones and since the fine tunning takes a while and the circuit is so close to be unstable is hard to be 100% satisfied. When I finished mine I didn't have any big issues so all I did was set all the knobs at noon and played. Basicaly my amp did what it was supposed to do, then curious me I started to read diferent topics about testing the amp in a way I never did like diming all the knobs without a cable plugged in and guess what I found? my amps squeals, then before that I tested tubes, biased one way then started moving wires and everytime u move something the amp slightly changes its sound...so at the end I wonder what is the correct way? some people kindly suggested using grid stoppers but then some other people says they don't use that and their amps are quiet...is all confusing!!!

[rooster]

OK, I have to drop in on this. Paulster, using your example the calculations are pretty simple using a bias probe like redshark has. If its just about measuring the voltage drop across the screen resistor and subtracting this current amount from the plate current to cathode found on the bias probe reading.

Which means a 1K screen resistors will typically drop 6 ma, all 470 ohm resistors 3 ma, etc. Is this the idea?

OK, just to test this thought, an amp like a 5F4 Super that has no screen and typically more voltage on the screen than the plate could use the exact reading of the bias probe to determine the plate current at the cathode. Is this correct?

Taken to an extreme, looking at a Silvertone 1484 with no screen resistors and yet a 112VDC reduction to that of the plates (470VDC), I would ask how is it that you would determine this cathode current number using a bias probe?

Well, to respond to redshark's question, no, I haven't heard this info before now. And it could be that I don't get around much. ? I have always treated the cathode reading to be the current on the plates - what the bias probe was meant to read to make adjustments against. I haven't heard about adjusting this number further by looking at the current across any resistance found on the screen resistor.

Is this what you are saying paulster? Not an argument, per se, but I have not heard this before. Where do I read this info? Thanks.