Using Resistors to Change B+

[The New Steve H]

Steve,
You've got to make up your mind about what you really want. If you want 250-0-250, that's what you should get. You say "Trainwreck" transformer. What TW? An Express uses a 585V supply. Rocket and Liverpool are EL84 amps, not EL34, so I am thinking you mean an Express tx. That's not a 500v tx.
Phil

I want a PT that will give me 500-525V.

The reason I said "Trainwreck transformer" is that this is what the seller calls it. I was looking at a Classictone. They claim their PT works with various TW amps.

Sorry about the EL84/EL34 confusion. I am starting to go crosseyed from studying this stuff.

[John_P_WI]

Steve,

Read and try to understand everything in this post by VacuumVoodoo:

https://tubeamparchive.com/viewtopic.ph ... highlight=

He is THE man of authority.

[The New Steve H]

I'm all over it. Thanks.

Before I slog through this, is this your polite way of saying Firestorm is crazy, or are you backing him up?

[Jana]

I'm the crazy one here!

[John_P_WI]

I'm all over it. Thanks.

Before I slog through this, is this your polite way of saying Firestorm is crazy, or are you backing him up?

No reference to Firestorm or anyone else intended. I do not know Firestorm but from reading his posts over the years believe that he is very grounded in his knowledge and approach in a good way.

I posted the link to VacuumVoodoo (Aleks) post as he is person who is a professional engineer and amp builder. As a fellow engineer, I hold him in very high regard from his posting days at powerscaling.com. In fact, I wish he would write a book on the how to's of proper amp design.... hint, hint Aleks.

[The New Steve H]

I've had a couple of problems which have made this mess harder to understand. I wanted to be able to treat tubes like resistors, so I would be able to figure out what should go on either side of them, but I don't even know if that works. Are their resistances relatively constant? Do they vary with applied voltage? I have no idea.

The other thing is that datasheets use weird terms that don't always seem to relate to tube amp jargon. I try to translate for myself, but it hasn't worked out too well. I look for the current that goes through the plates, and I see things like "I" with a subscript "a." Is that it? Danged if I know.

I'm looking at a 6BM8 datasheet right now. There is no mention of internal resistance. There are two terms expressed in ohms. One is "plate load," and the other is "plate resistance." I don't know what those things mean.

I'd like to be able to look at this sheet and know how much resistance there is when current goes from the B+ to the cathode, but I don't see it. The schematic I have says there are about 225 volts between the plate and cathode, but I don't know the current.

I think if I could spend half an hour with an engineer and ask some simple questions, this stuff would make a lot more sense.

[Jana]

Steve,
Are you frustrated enough to rip it apart and start over? If so, I have a suggestion for you. Keep in mind, I'm the crazy one here.

You want a 15 watt amp that is as close to the 59 Bassman circuit as possible. You love that topology--not a bad one to love. So, here's what you do: use your existing PT, one of the OT and as many other parts as you can salvage and build another 59 Bassman just like you did last time. Everything the same. The only difference is, use 6V6 tubes in this one. The voltages from your PT should match up just about perfect for this. You are using a Deluxe OT, right? Again--just right for 6V6. This should give you about an 18 watt amp. If you want a Hi/Lo power switch, add a switch on the back panel to switch from pentode to triode--do it the way Marshall does it by switching the screen grid resistors from the screen supply to the plates. This should drop the amp down to about 7 watts or so. Use fixed bias just like in your other Bassman build. You will have to tweak the bias for the 6V6 tubes, but that is easy.

[The New Steve H]

It's a good idea, but I am not that fond of 6V6 tubes.

[Jana]

Same thing with a pair of EL84? Or did those stiff you on a date too?

[Firestorm]

Before I slog through this, is this your polite way of saying Firestorm is crazy, or are you backing him up?

No, Jana's the crazy one here

I'm just saying that the 6BM8 35mA "rating" is only a typical use example assuming 200V B+ in Class A. That isn't what you're building, so you need to look at your intended B+ and calculate current at idle and current at max signal to see what ballpark you're in. I don't know what the rest of your circuit looks like, but unless you're driving the tubes to their maximum output at your intended B+ full time, you probably get away with 100mA if the PT is rated conservatively. For comparison's sake, consider that the Tweed Deluxe ran two 14W tubes off a transformer rated for about 80mA.

[diagrammatiks]

Steve I thought you said that your power supply was sagging too much and giving you lower voltages?

If you look at the original little wing schematic.

The PT is 260vac/520vct@100ma.

That gives you a maximun of around 360vdc off that winding.

It looks like a 5y3 will drop about 60 volts while supplying 125ma.

So that's 300vdc after the rectifier. That means if you have no tubes plugged in the amp you should be reading about 300 at the rectifier cathode.

Looking at the little wing schematic - under full load he gets 275 at the rectifier cathode. That's about 25vdc drop under load. That's all good.

Here's who to figure out what the load is for the plates...

Look at the cathode of the power tubes. The voltage given there is 22.9 with a 420ohm resistor. Using ohm's law we can figure out that the tubes are drawing 55ma idle for 2 or 22.5ma idle for each tube.

If you had 4 tubes sharing a 210ohm resistor you would expect to have 22.5 *4 or 90ma as the idle draw for the plates. .

then you calculate the screen draw. Looking at the schematic again...the voltage is 276vdc before the screen resistor and 250vdc after the screen grid resistor. The drop is 25vdc across 4.7k

That puts the draw at about 5ma per screen idle. With two tubes the screen draw would be 10ma and with 4 tubes it would be 20ma.

So for the 2 tube version the power tube draw is about 60ma and the 4 tube version is bout 120ma.

Now you can calculate the preamp idle draw.

From the schematic node B is 276 and node C is 270 across a 1k resistor.
That's about 6ma for the preamp as given.

That means that idle current for

2 tubes is 66ma
4 tubes is 126ma

Now you just need to give some leeway for swing current. Let's just estimate 50 percent of idle. One half of 60 is 30 plus 6 makes 96. The schematic you posted uses a 100ma rated transformer.

4 tubes is 126. One half of 120 is 60 plus 6 is 186ma as a nice and safe number for your 4 tuber.

No need for jargon, fancy maths or datasheets man.
That schematic you posted as all the numbers you need to figure out anything you could ever want about that amp.

edit: man I shouldn't be allowed to do maths in my head. half of 54 is 27 not 22.5

I rounded to 25 later so numbers should be ok.

[Phil_S]

I've had a couple of problems which have made this mess harder to understand.

Like any problem, break it into smaller parts and try to understand it.

I wanted to be able to treat tubes like resistors, so I would be able to figure out what should go on either side of them, but I don't even know if that works. Are their resistances relatively constant? Do they vary with applied voltage? I have no idea.

I think maybe you are getting the idea. Tubes have multiple attachment points and respond to both what's connected and the inputs. They are dynamic. You can't treat them like resistors. Look up "drawing a load line." Try the Valve Wizard.

The other thing is that datasheets use weird terms that don't always seem to relate to tube amp jargon. I try to translate for myself, but it hasn't worked out too well. I look for the current that goes through the plates, and I see things like "I" with a subscript "a." Is that it? Danged if I know.

It is simpler than it looks. The abbreviations are standard electronics jargon. I bet you can find it in Wiki.
"I" is the symbol for current in Amperes (Amps).
"a" is the symbol for anode (also called "plate"); you'll find more typically Brits call it anode and Yanks call it plate.
"Ia" is anode (plate) current
OK, now you know. Other stuff:
"R" for resistance in Ohms
"k" for Cathode
"g" for Grid
"g1" for the Grid in a pentode (the signal input point)
"g2" for the Screen Grid (the anode helper) in a pentode
"g3" for the Suppressor Grid (normally attached or strapped to the cathode) in a pentode
So, for example, Rk is the value of the cathode resistor and Ik is the plate "load" resistor.
Now you know more about basic tube parts.

I'm looking at a 6BM8 datasheet right now. There is no mention of internal resistance. There are two terms expressed in ohms. One is "plate load," and the other is "plate resistance." I don't know what those things mean.

Plate load is what your output transformer primary needs to "show" the plate in terms output impedance to work correctly. This is a fixed value at a certain plate voltage. You draw a load line to see how plate load changes as you change plate voltage.

Plate resistance is the internal resistance of the plate structure itself and is useful in certain calculations. It is also useful in determining if one tube is similar to another.

I'd like to be able to look at this sheet and know how much resistance there is when current goes from the B+ to the cathode, but I don't see it. The schematic I have says there are about 225 volts between the plate and cathode, but I don't know the current.

You can probably figure this. If plate voltage (Va) is 250 and cathode voltage (Vk) is 25, then you have 225V between them. To get the current you've got to measure something while the tube is active. The easiest thing to do is put a 1 ohm resistor between the cathode and ground, and clip your meter to each side of that 1 ohm resistor. Meter for DC volts. This will tell you the voltage drop. Apply Ohm's Law, V=I*R to solve for I. So, I = V/R. Since R = 1 in this case, V=I. It is a safe way to determine current. Measuring I directly with the Amp setting on your meter is kill you dead dangerous and not recommended for amateurs who aren't sure what they are doing. The other way to do this is to know the R of the half of the output transformer supplying the plate and use that R value along with plate (anode) voltage to solve for I.

If you are just looking for the theoretical, use the values on the data sheet. Usually there are a few fixed choices within the data. You can try to guess based on what's there if your particular voltage isn't listed.

I think if I could spend half an hour with an engineer and ask some simple questions, this stuff would make a lot more sense.

They don't teach this stuff to engineers any more. You need to find an older guy -- someone who is at least 60 years old. Go to a hamfest and see if you can charm one of those guys into sitting with you for a bit. Get up before you go to bed, though. The hamfest crowd likes to start things up as soon as the sun comes up. I got to one at around 11AM and it was all but cleared out!

[The New Steve H]

I decided to do some math. I treated the preamp tubes like resistors and drew an equivalent circuit. I came up with 3.5 mA per side of V1. Either that or 1.75 mA; I don't have the paper in front of me. I used the anode-cathode voltage drop, and I came up with about 85K for the resistance of the tube. Putting all the values back in, it looked like I needed 68K resistors to replace the 100K jobs on the V1 plates, to raise the voltage from around 135 to 150.

Then I turned the amp on and checked the voltages on V1, and they were already nearly ideal (near 150).

Super irritating.

Then I worked on the amp for a while, without doing anything to the preamp. I just cleaned up the wiring in the output area. I checked V1 again, and the voltages were down to about 135.

Here's what's happening now. The amp starts to warm up, and for a few seconds, it seems to work. Then the voltages on the power caps drops. In the end, the plate and screen voltages become almost equal, at about 250.

Somewhere, current is finding its way out of the output tubes when they warm up fully. Unless I'm going crazy. But I don't see how that can happen.

I keep checking the tube connections, but they look fine.

Before the voltage drop, everything looks good. I checked voltages all over the place.

The fact that the amp works for a few seconds suggests that the PT is at least adequate for testing purposes. If it couldn't push the tubes, it wouldn't start strong and then fade.

I am wondering if there is something about these tubes I don't know. I dismissed the cathode/heater bias thing, but for all I know, it's coming into play. I can't find any literature which explains it. All I know is that under SOME circumstances, the cathode and heaters have to be within 100 volts, and with the cathode at 24 and the heaters at +/- 3.25, I do not see how this can matter.

[diagrammatiks]

the tubes are cathode biased.

how long is it taking until the voltages drop?

you'd have about 2-3 seconds before the tubes hit idle current draw from warmup.

Although, you're getting 250 on the plates?

that should be about how much drop you'd get.

i think you should for right now, pull two tubes and one output transformer and see if you can get it working.

[The New Steve H]

I've run it on two tubes with an OT disconnected. If that was the problem, it would have been working last week. These tubes are wacky. Where else could they be sending current?