Extrapolating loaded voltage on PT

[Tyrannocaster]

I am finally starting a project I have been waiting for parts for. I have a PT from an old Hammond tone cabinet and I have just measured its voltage unloaded. I get 710 across both leads, or 355 per side. I had planned to use a SS bridge rectifier with this transformer and push a pair of 6L6s in cathode bias mode (Fender Tweed Pro 5E5) and I am wondering if this is not going to be too much, however since this reading was taken with no load, is there a figure of thumb I can use to guess the drop after it powers up the 6L6s and three 12AX7s (I guess the preamp tubes don't really matter that much except for their heater current and I have plenty of heater capacity)?

The PT originally was used with a couple of 6V6s and a ton of preamp tubes and according to the schematic delivered something like 310 after rectification to them, which I why I thought the SS rectifier would be good with the 6L6s for the higher voltage that a Fender Tweed Pro wants. But line voltages are higher now than they were then and of course there is no load on my tranny now as I read the voltages.

Seems like a question that someone with lots of hands-on experience should be able to answer easily, but I am not one of those people. 

[sluckey]

710 across both leads, or 355 per side. I had planned to use a SS bridge rectifier with this transformer

A bridge will give 710 x 1.414 = 1000v unloaded B+. Better stick with a conventional 5U4 or 5AR4.

A conventional rectifier will give 500v unloaded B+. The B+ will drop when loaded with the tubes. How much depends on how hot you bias the tubes, the drop across the rectifier tube, and the internal resistance of the PT winding.

[Tyrannocaster]

That's what I was afraid of. I can't imagine how they used this PT with a pair of cathode biased 6V6s!

EDIT: I just realized I posted this in the wrong forum. If somebody could move it to the Tweaks forum I'd appreciate it. It doesn't belong here.

EDIT 2: Looking at the schematic for the Hammond AO-29 (which is the model I took the PT out of) it shows 320 volts on each side before the 5U4 rectifier tube. That's why I had thought the bridge should be about right for the 6L6s. Trying to put another tube in this ting would be a problem; I may be in for a zener drop. I don't like those, but they work if it's the only thing you can think of to drop the voltage.

[sluckey]

There are two versions of the AO-29. The early one uses a 380-0-380 PT and has a field coil speaker. B+ is 320, but the field coil is in the center tap of the PT HT winding and it has -100v across it (like having a 100v zener). So that's up to 410vdc. And there are several bleeder resistors to help drop the B+. Quite a load.

The later version shows a 320-0-320 PT and a more familiar power supply.

So, which do you have? Can you post your schematic?

PS... I'll move this thread to the tweaks forum.

[Tyrannocaster]

Thanks. Beyond the empty chassis (amp is totally gutted) and the trannies I don't have any information on the amp. I didn't know there were two versions; I suppose it would be my luck to get the one with a 380 volt supply but I don't know how I would tell other than by measuring it. There's no organ, speaker, cabinet, anything at all to look at. The OT on the chassis is pretty small, not even as big as the one in my '55 Tweed Deluxe. But there were other items plugged in, as you see from the schematics. The full number on the chassis is AO-29-13 and perhaps the 13 designates whatever difference there might be. The only schematic I have is the one from the Captain Foldback site (Mid-Sixties w/reverb) and it's quite low-res - I can't read the values of the resistors in the power supply, for example. I believe mine is that vintage because that is the vintage that the ones I can find are with the same full model number.

So they must have been doing something like what you described to get the voltage down for a pair of cathode biased 6V6s. And that puny OT wouldn't take much power at all. Guess that's where Steve Winwood was getting part of his distortion on I'm A Man.

I did find an AO-29-13 on ebay that needs a power transformer, LOL: http://www.ebay.com/itm/Hammond-M-100-Organ-Amp-Model-AO-29-13-Chassis-Needs-Power-Transformer-/200760659336

[sluckey]

Back to your original post... You measured 710 (355-0-355) on the HT winding. That's unloaded. And that's with today's higher line voltage. That could very well be the later PT that put out 320-0-320.

I would use it on a pair of 6L6s and a 5AR4 rectifier. Once loaded I bet the B+ will be somewhere around 450, a very real number for a pair of 6L6s.

[Tyrannocaster]

The problem is the tube socket for the rectifier - the chassis is all ready and I can't put another one in. It was very difficult cutting the holes for the ones I did put in and I can't figure out how I would fasten the chassis down now that everything is in place - before, I actually screwed it onto a big sheet of plywood so I could get enough leverage to use the chassis punch. The preamp tubes weren't too bad, but those octals were quite a lot of work. But worse than that, there really isn't any place to put another octal tube.

My mistake was in thinking the PT would work for this purpose and then getting the chassis ready. I will have to decide where to go from here; I'm stumped for the moment - more precisely, I'm bummed at the moment, but I will bounce back. 

[phsyconoodler]

500v will work too.Just bias accordingly or use a mosfet or zeners to get the voltage down a bit.Easy to do.

[Tyrannocaster]

I could make a Plexi variant but I don't trust this PT for the current it would need for a pair of fixed bias 6L6s. That's why I was planning on a Tweed Pro - I know it can cope with that, but the voltage is just so high. I have used zeners before (although never with a SS rectifier) and found the heat to be a problem - I was using several small ones in series and the solder connections actually came apart from the heat - ugh. If I could figure out how to put another octal socket in there I'd be home free - I could even use a Copper Cap for a custom voltage if I wanted (but I think those are just more zeners in addition to a regular diode rectifier - anybody confirm that?).

500v will work too.Just bias accordingly or use a mosfet or zeners to get the voltage down a bit.

I don't think you can bias a cathode biased Tweed Pro for that much voltage.

EDIT: How far can you go using a resistor to drop voltage instead of a zener? I have read that doing so gives more sag like a tube rectified amp, and that would be appropriate for a Tweed Pro. I realize it might need to be a rather hefty resistor (with good dissipation avenues), but is it even possible to drop enough voltage to make this PT workable for this application?

EDIT #2: Okay, I have been looking at the layout and I can move some things around. Then there is an area where I could put in an octal socket. It's smack dab between the OT and the PT; there is exactly two inches between their end bells/frames. So there is enough room physically, but is there any potential problem with putting the tube there? I have seen some pretty odd setups so I would tend to think this is okay, but I would rather ask so I don't get any more nasty surprises.

[Tyrannocaster]

Bump.

Guys, I have put an octal socket in now and I'm going to continue with the build. If absolutely necessary I will either use a Copper Cap or add zeners to the center tap (now that I have one - I had taped it off for the bridge rectifier) but I would like to know how high you can go with 6L6GCs in cathode bias. If sluckey is right and I end up with 450 volts on the 6L6s can they take that in cathode bias mode with a different resistor?

[HotBluePlates]

... I would like to know how high you can go with 6L6GCs in cathode bias. ...

450v B+ is perfectly reasonable for cathode-bias 6L6GC. Or probably even 5881, 6L6GB, maybe metal 6L6.

But let's assume a 6L6GC, 30w dissipation rating. 30w/450v = 66mA at idle.

Could this be run class A? Cathode bias doesn't necessarily dictate class A, but working far into class AB works much better with fixed bias. Let's assume class A.

If the peak current is double idle current, the peak for this tube is 132mA. When running class A, the load to one output tube is 1/2 the plate-to-plate load. Let's guess a 6.6kΩ OT; the load to one tube is 3300Ω. 0.132A*3300Ω = 435v. The 6L6 won't be able to pull its plate this low (down to 15v on the plate), but will come close, maybe to 50v (400v peak).

400v peak * 0.7071 = 282v RMS. 0.132A * .7071 = 0.093A RMS. 282v * 0.093A = 26w RMS as a guess of output power.

The G.E. 6L6GC data sheet has plate curves for 400v on the screen on page 7. If you draw a vertical line 450v plate, and a horizontal line for approx 66mA, they will meet around midway between the grid line for -30v and -40v. Your screen voltage might be a bit over 400v, which will increase tube current somewhat.

Let's say we guess the bias should be -37v. The idle current for two tubes will be 0.066A * 2 = 0.132A. 37v/0.132A =  280Ω. 37v*0.132A = 4.88w A 10w resistor should work.

That seems like an odd value, but Mouser has Dale wirewound resistors at 280Ω and 300Ω (non-stocked, maybe pick a different brand or different source). 300Ω * 0.132A = 39.6v, and may be a better choice for 450v on the screen. 39.6v * 0.132A = 5.22w.

The best option is to buy several values around this range (a bit above 300Ω down to 280Ω) to be able to swap a higher value if you find a lower one allows too much idle current.

So yes, it can be done. You could probably also use a 4k plate-to-plate OT if that's what you have on hand.

450v - 37v bias = 413v plate to cathode. 413v * .066A = 27.5w. That's almost low enough to run 6L6GB or other 25w rated 6L6 variants.

[Tyrannocaster]

Wow, HBP, my head is spinning from that math.  But your very first sentence was the takeaway for me, so thanks a big bunch.

As far as the resistor goes, I will get several to try. BTW. I will be using separate resistors for each 6L6 rather than one common one (adjusting the value for individuals, of course).

Thanks again. That takes a load off my mind.

[HotBluePlates]

Then you'll need pairs at 560Ω up to 600-620Ω @5w. Mouser has those from various brands in stock.

FWIW, I have an amp build now with 807's (25w 6L6GB with a top cap), cathode biased, with 430v on the plates.

... I will be using separate resistors for each 6L6 rather than one common one (adjusting the value for individuals, of course). ...

You can do that, but each tube will settle in on a somewhat different value of bias depending on it's transconductance. So you don't gain much by using separate resistors.

My build uses a common cathode resistance for the output tubes, but does replace a portion of that with a 100Ω pot to trim the idle current. There is an additional pot which adjusts the ground reference between the output tubes to act as a bias balance. Look at the output stage of a Williamson amp for the general method.

[PRR]

> You could probably also use a 4k plate-to-plate OT

I would keep the OT impedance similar to the original. This impedance indirectly sets the full-load current.

So if the Hammod donor was a pair of 6V6 or 6BQ5, typically 8K, use 8K. If quad-6V6, indeed 4K is fine.

[HotBluePlates]

Yeah, I don't know the reasoning other than feared too-high B+, but he was thinking a pair of 6L6.

But, yes... If the PT came from a Hammond with a pair of 6V6, you want to use an OT for a pair of 6V6, as well as a similar-to-original output section.

Production amplifiers don't usually build in a lot of reserve for using bigger tubes. That's generally spending possible profit on a better-than-needed part. Transformers seem like a good place where you either make your money, or build for a loss (production-wise).

[Tyrannocaster]

I have a pretty hefty 5.5k OT for the amp and I believe it should more than handle anything a pair of cathode biased 6L6s could put out. I don't know what it came out of but it worked out to 8x690 (10.25 volts in gave .39 volts out. 10.25 / .39 = 690, which is my impedance ratio), or 5.5k at 8 ohms if my math was right. (I did the calculations a while ago.) I think the 6L6s should be fine with the PT from the Hammond, given how much it was supplying in the AO-29 (lots of other tubes) and that the 6L6s will be cathode biased. If they put out 26 watts that's still a lot less than they might do in fixed bias mode and not THAT much more than a pair of cranked 6V6s. Since the OT is more than adequate my only worry was the voltage on the 6L6s. I know that traditionally the old Tweed Fenders did not go that high (the 5E5 shows 390 IIRC) and the tube data on the 6L6 in my RCA manual stops at 250 volts in class A push pull. Obviously, I know they can go a lot higher than that, I just didn't know how much.

I was going to use my EL37s in this amp but after you mentioned the heater current issue with them I decided not to since I don't know how much current the Hammond can actually sustain. But I'm sure it can deal with two cathode biased 6L6s, a tube rectifier and three 12AX7s. (The 5E5 didn't have tremolo but I am putting in a tube for that, the circuit to be added after I get the amp working satisfactorily. Uh...a Promolo? A Vibropro?  

[HotBluePlates]

Well, the concerns are heater current and B+ current availability. What Hammond did the chassis/PT come from? Is there a schematic you can point to?

A lot of little tubes may not = B+ current availability for a big output stage, unless it's a LOT of little tubes. Or maybe it will be just fine. A schematic would probably tell the tale, especially one with voltages indicated.

[Tyrannocaster]

Well, the concerns are heater current and B+ current availability. What Hammond did the chassis/PT come from? Is there a schematic you can point to?

A lot of little tubes may not = B+ current availability for a big output stage, unless it's a LOT of little tubes. Or maybe it will be just fine. A schematic would probably tell the tale, especially one with voltages indicated.

As mentioned in the original post, it's from a Hammond AO-29-13, which was used in the M-100. The only schematic I can find is on the Captain Foldback site but it's zipped ( http://captain-foldback.com/Hammond_sub/schematics/M100sch.zip ). At any rate, since the cathode biased 6L6s don't put out that much more wattage than a pair of screaming 6V6s, and there are more than enough tubes in that organ amp to provide heater current for everything I'm not particularly worried about that part of things. Voltage is 320 (originally, with the old lower wall voltages) on the schematic (as mentioned earlier) but I am getting more like 350.

Like I said, I wouldn't do this if the tubes were fixed bias. And I have always found that organ amp PTs have way more capacity than you expect - well, at least the ones I have run into do; I've only had trannies from Hammonds and Baldwins, though. On the other hand, their OTs are often incredibly wimpy.

[HotBluePlates]

Oops... I saw that after posting.

So the schematic you linked shows a B+ node of 320v, 500Ω resistor, then a 295v node. 25v/500Ω = 50mA for the screens/preamp. Yes, you're probably good to go with the 6L6's.

Like I said, I wouldn't do this if the tubes were fixed bias.

Well, the original circuit is fixed-bias. A back-bias arrangement provides fixed-bias and wastes a little voltage. And Hammond used a series resistor ahead of the first filter cap to approximate a choke-input supply and lower the B+. I think there's a little note by the input resistor saying "140mA", but I can't make it out under magnification. They never should've saved the scan as a jpeg.

[sluckey]

Here's the early version. Big diff in the power supply. Maybe this will help you determine which one you have.

http://hammondb3organ.net/schematics/hammond/organs/m3/AO-29-fc.pdf

[Tyrannocaster]

Sluckey, I don't know how I would tell them apart since all I have is the two trannies and a totally gutted chassis. The chassis are physically identical in the schems except for the slightly different outlines of the PTs on the schematic and I don't know if those are accurate representations. If the earlier one put out 380 volts then mine is actually low, as with today's wall voltage I would expect even more than I am getting so I am staying with the idea that it is the later version. This is corroborated by the fact that all the references I can find to AO-29-13 reference the Sixties model with the lower voltage PT.

At least you can read the schematic for the earlier one, though; that other one is so blurry it's hard to make out a lot of stuff, as HBP said.