Planning my BA-6A

[Brian Roth]

I've been following this thread with some interest, mainly related to the power supply issues.  I decided to do some "what if?" sims with the Duncan PSUD2 app, largely to see how it compares to "reality".

I guess the original RCA HT secondary winding had relatively high series resistance.  At least that's the only way I come up with a final B+ voltage of 285 VDC with 100 or 125 mA load at the output of the "pi" filter (and the series resistance of the choke as shown in the RCA schematic).

I see at the beginning of thi thread that a Hammond xfmr with a much higher secondary voltage than the RCA original is being proposed.  Which model Hammond is under consideration (since Hammond publishes series resistance specs)?

Regardless, one thing struck me, and that is the commonly available 450 VDC rating will likely be too low for the 40 mFd filter caps if using the proposed higher voltage iron, and with the proposed additional dropping resistor AFTER the "pi" filter.

Best,

Bri

 

[abbey road d enfer]

Good question, Brian.
I took letterbeacon's word for granted when he said 250V, but I see that Hammond have a model 369GX that is rated at 450V center-tapped, that's good for 318Vdc @ 100mA, according to this chart
http://www.londonpower.com/hammond/classic1.htm
Seems to me it would fit nicely and get rid of the additional drop resistor.

 

[letterbeacon]

I've been going from the table on the Hammond website here: http://www.hammondmfg.com/300series.htm where the specs of the transformers aren't as comprehensive, weirdly.

The 369GX unfortunately doesn't have a 5V tap for the 5R4GY rectifier that I'd like to use.  Looking at that London Power page, how about the 370DX/Z?  That's 388Vdc @ 138mAdc at half bridge full-wave rectification.

 

[abbey road d enfer]

I've been going from the table on the Hammond website here: http://www.hammondmfg.com/300series.htm where the specs of the transformers aren't as comprehensive, weirdly.

The 369GX unfortunately doesn't have a 5V tap for the 5R4GY rectifier that I'd like to use.  Looking at that London Power page, how about the 370DX/Z?  That's 388Vdc @ 138mAdc at half bridge full-wave rectification.

370DX seems overspec'd at 2x275V.
369HX with 2x240Vac seems perfect, but a tad overrated at 120VA.
370CAX seems to have the right combo of features; 2x250Vac may need a voltage drop resistor.
I had not given much thought about the use of a vacuum rectifier. R4 has about 30V drop at 100mA; that may just be what you need.
Do you still plan to use a separate xfmr for the 6.3 heaters?
If so you could use 369GX with a 0R68 resistor in series for the 5V.

 

[letterbeacon]

Do you still plan to use a separate xfmr for the 6.3 heaters?
If so you could use 369GX with a 0R68 resistor in series for the 5V.

Yes I'm still planning on using a separate xfmr for the heaters, so this looks like a plan.  Would I still be able to tap off 3.15VAC before the resistor to feed the balance circuit?

According to the Hammond website it says the 369GX is 450CT @ 75ma.  I thought I needed 100ma?

369HX with 2x240Vac seems perfect, but a tad overrated at 120VA.

It doesn't matter too much if it's overrated though, does it?  Apart from the financial implications!

 

[abbey road d enfer]

Do you still plan to use a separate xfmr for the 6.3 heaters?
If so you could use 369GX with a 0R68 resistor in series for the 5V.

Yes I'm still planning on using a separate xfmr for the heaters, so this looks like a plan.  Would I still be able to tap off 3.15VAC before the resistor to feed the balance circuit?

Not sure I fully understand your question. You don't want to balance the winding that's used for the valve rectifier. My suggestion implied that the 6.3 winding of the 369GX would be used only for the rectifier, which draws 2amps. Anyway, the rectifier's heater winding is also the B+, so you don't want to fiddle with it.

According to the Hammond website it says the 369GX is 450CT @ 75ma.  I thought I needed 100ma?

Which makes it good for 225V @150mA or 320Vdc @100mA.

369HX with 2x240Vac seems perfect, but a tad overrated at 120VA.

It doesn't matter too much if it's overrated though, does it?  Apart from the financial implications!

Just more bulky. Then it seems like a pity not to use this 6A 6.3V winding...

 

[letterbeacon]

Yes I'm still planning on using a separate xfmr for the heaters, so this looks like a plan.  Would I still be able to tap off 3.15VAC before the resistor to feed the balance circuit?

Not sure I fully understand your question. You don't want to balance the winding that's used for the valve rectifier. My suggestion implied that the 6.3 winding of the 369GX would be used only for the rectifier, which draws 2amps. Anyway, the rectifier's heater winding is also the B+, so you don't want to fiddle with it.

On the original schematic I've attached, one side of the 6.3 winding is used to feed the 6SK7 balance circuit to enable to user to balance the tubes before operation.  If I was to put a resistor in series to drop the 6.3v to 5v to feed the rectifier, would I still be able to use one half of the 6.3v to feed that balance circuit?  Perhaps tap the 6.3 winding before the series resistor.

 

[abbey road d enfer]

On the original schematic I've attached, one side of the 6.3 winding is used to feed the 6SK7 balance circuit to enable to user to balance the tubes before operation.  If I was to put a resistor in series to drop the 6.3v to 5v to feed the rectifier, would I still be able to use one half of the 6.3v to feed that balance circuit?  Perhaps tap the 6.3 winding before the series resistor.

No, big misunderstanding here. The center-tap of the 6.3 winding doesn't feed the balance circuit, rather the contrary. The bias voltage created by the voltage divider from the regulated supply "feeds" the 6.3V winding. In fact, feeding is not adequate since there is no current drawn, it's a polarisation, which creates an electrostatic field in the vacuum tubes, between filament and cathode.
Note how the rectifier heater is connected to B+. You could use a 6.3V winding with adequate series resistor to power the 5V filament; 1.3V drop with 2Amps computes at 0.65ohm - closest value 0.68ohm, dissipation 2.6W.
Now, if  the heater voltage for the other tubes is DC, you don't need to elevate it. At first, with my incorrect schemo, I thought it may be needed in order to minimise Vk-h, but now, it's clear the bias is there only for improving noise performance of AC heaters.

 

[letterbeacon]

I think we're talking about slightly different things here.  I don't mean the heater bias, I mean the section I've highlighted in red in the attached picture (The full schematic is here).

This section, as I understand it, is a way of balancing the 6SK7 tubes.  Using the rotary switch you can apply 3.15VAC to the grids of the 6SK7.  You then adjust the balance pots until the meter reads zero.  On the original schematic the 3.15VAC is obtained from one half of the 6.3 heater winding.

My question is if I'm using the Hammond xfmr and therefore  dropping the 6.3VAC to 5VAC to feed the heater.  Am I still able to tap the 3.15VAC from the same 6.3 winding without causing problems with the rectifiers's heater supply.

 

[abbey road d enfer]

My question is if I'm using the Hammond xfmr and therefore  dropping the 6.3VAC to 5VAC to feed the heater.  Am I still able to tap the 3.15VAC from the same 6.3 winding without causing problems with the rectifiers's heater supply.

OK, I follow you now. Yes you would still be able to tap the winding for the balance tone, BUT this would be refernced to B+ instead of the 20-30V bias, so you would need to be careful with the proper working voltage for the caps (C12 & C13 on the original RCA schemo) and you should provide a discharge path for these caps where they are connected to the rotary (1-10Megs to ground).

 

[abbey road d enfer]

Now I'm a tad puzzled because your choice of a vacuum rectifier changes the requirements for the HV winding. The original schemo indicates 320Vac (voltage chart says 315Vac) but the unreg voltage of 285V seems to indicate quite a high voltage drop in the rectifier AND xfmr windings. Peak value for 315Vac is 440V, that would infer 150V loss in the rectifier + DCr...seems too high to me. Discrepancy may be due to the fact that voltages are generally measured with an average-value meter compensated for rms indication; since the secondary voltage is highly distorted, the compensation doesn't really work.
I guess Brian may chime in, since he has probably a fresher knowledge of valve rectifiers (I have ceased using valve rectifiers once and for all in 1967).

 

[letterbeacon]

I see, I think that's what Brian meant earlier at the top of the page.

I guess the original RCA HT secondary winding had relatively high series resistance.  At least that's the only way I come up with a final B+ voltage of 285 VDC with 100 or 125 mA load at the output of the "pi" filter (and the series resistance of the choke as shown in the RCA schematic).

Perhaps I should just build it as is and then put a dropping resistor in there once I know what the circuit supplies in the real world.

 

[analag]

Didn't read the entire thread, but a great way of dropping B+ and improving regulation at the same time is to remove C16 and probably use a higher inductance choke. Aka inductance input filter...the rectifier tube will live longer too.

 

[Brian Roth]

I guess Brian may chime in, since he has probably a fresher knowledge of valve rectifiers (I have ceased using valve rectifiers once and for all in 1967).

In no way do I claim to be an expert on the topic of tube rectifier circuits!  <g>  That's why I fall back on PSUD2 when I see something "interesting".  Sims with PSUD2 usually come pretty close to what is shown in the Olde schematics I run across.

I did find some interesting stuff in my Olde RCA manual (RC-28), from 1971, so I scanned it and am posting it here.  The example circuit 1 and filter 1 are the same toplogy as found in the BA-6A.  The specified transformer was a 300-0-300 unit, which is pretty close to the specs shown in the BA-6A schematics.

RCA apparently was "pimping" their 5BC3 rectifier at the time.  I compared the specs between the 5BC3 and the 5R4 and found the BC3 was a bit beefier than the R4 in terms of max. current, and hence the BC3 has lower plate impedance...implying a lower voltage drop.

As you can see, with a pi filter and 120 mA load, the output voltage "sags" to 320 VDC.  That makes what is shown in the BA-6A schemo seem somewhat believable after all.

I couldn't find any specs for the Stancor iron listed in the RCA manual, nor does PSUD2 have a model for the 5BC3.  So, no way for me to sim the RCA manual's circuit with PSUD2.

Best,

Bri

 

[Brian Roth]

I see, I think that's what Brian meant earlier at the top of the page.

I guess the original RCA HT secondary winding had relatively high series resistance.  At least that's the only way I come up with a final B+ voltage of 285 VDC with 100 or 125 mA load at the output of the "pi" filter (and the series resistance of the choke as shown in the RCA schematic).

Perhaps I should just build it as is and then put a dropping resistor in there once I know what the circuit supplies in the real world.

If using the Hammond xfmr with the much higher secondary voltage ratings, be sure you "up" the voltage rating for the two filter caps!

Best,

Bri

 

[CJ]

Hammond will wind you a custom trans for a few more dollars, not that bad at all.

i think they have many blueprints for all manner of winding configurations that are not listed in the catalog, so if you order a custom, chances are they have already built it for somebody else,

does sowter make a pwr trans for this unit?

let me go see...

a 5 volt heater winding might have better isolation voltage between ground than a regular 6.3 winding, since the engineers know that 5 volts = HV rectifier, an additional wrapper or 2 of insulation might be used.


FYI  a 2 volt rect winding has mega insulation, since it is used for a RF transmitter etc, the RF will bridge to ground easier than 60 cps.


ok, sowter sells in mid and out, but no pwr.

i bet Brian would be interested in adding the pwr trans to his vintage line, you might give him a call.

 

[letterbeacon]

Thank you Brian and CJ.

CJ - I've emailed Hammond about a custom transformer, I don't expect to hear back from them until the new year, but there is a note on their website that says they do custom transformers with a minimum order of $500!  If I'm just adding a winding to an existing transformer it might be different I guess.

I also had a look at Brian Sowter's site and according to the custom section, I could be paying about £140, which also is a little out of my price range.  I'll call him in the new year and see what he says though.

I'm afraid to say I'm getting a bit lost here!  If anyone can point me in the direction of a good website on the theory of power transformers I'd be grateful.  I've been Googling but I haven't found anything which much detail!

It seems that all the off the shelf PTs that Hammond make (http://www.hammondmfg.com/300series.htm) output way more voltage than I need, even with a dropping resistor in there, unless I rely on the 150V getting lost somewhere!

 

[emrr]

Yes, the majority of modern PT offerings seem to be aimed at guitar amp or hifi design, with far too high a current spec for a given voltage, or far too high a voltage for the current spec.  Over here I'd probably call Edcor.  Dig through their giant list of PT's if you haven't.  I'm sure it'll still take more than one to get every voltage. 

 

[letterbeacon]

It looks like Edcor might be a good call.  This looks like it might be the closest to what I want:

http://www.edcorusa.com/products/602-xpwr012_240.aspx - 350vac @ 100mA, 6.3vac @ 2A, 5v @ 2A.

I'd still need another transformer for the DC heaters, but that was the plan anyway.  Does this look about right?

Edit - the shipping costs more than the transformer - but still works out cheaper than a Hammond!  Is there a quality difference between a Hammond and an Edcor do you think?

 

[abbey road d enfer]

It looks like Edcor might be a good call.  This looks like it might be the closest to what I want:

http://www.edcorusa.com/products/602-xpwr012_240.aspx - 350vac @ 100mA, 6.3vac @ 2A, 5v @ 2A.

I'd still need another transformer for the DC heaters, but that was the plan anyway.  Does this look about right?

Edit - the shipping costs more than the transformer - but still works out cheaper than a Hammond!  Is there a quality difference between a Hammond and an Edcor do you think?

2x175Vac is definitely not enough.