Edcor Custom PT?

[Willabe]

I'm working on a new scratch build. Parallel SE Champ with Power Scaling and 2 tube reverb, etc.

Decided to call Edcore for a bid on a custom PT. Spoke with an engineer there, told him what I wanted the rectified (SS FWFB)  B+ dcv to end up at (360dcv) and he said I had to give him an ACv for the B+ secondary. He said they don't do that because he didn't know if the B+ was going to be resistance, capacitor or choke loaded. I said it was cap loaded. Said he still wouldn't do it.      

I said your a transformer engineer, you know the math to make it come out at my target dcvI said if the secondary had a CT I could give him an ACV but I don't know how to figure it for a secondary with no CT. He said there's lots of equations in books but still said he needed a acv because he didn't know how I was going to build my PSU and how I was going to use it.            

I thought I'd go with 6.3 and 12.6 for heaters because then I can use a DC stand off on the 12_ _7 and pull a 5dcv from the 6.3 for any relays (got Doug's new boards) with out a problem from the dc stand off voltage. Or I might go with only the 12.6 and add a Hammond 224 series PT for the relay PSU. Very small and it's simi toride, so is very quite.

Here's the specs I gave him;

120vac primary

B+ secondary, 360vdc @120mA, no CT, SS FWFB

6.3vac CT @ 2A (for power tubes)      12.6vac CT @ .7mA (for 12_ _7)

Now what?????

I'm sure if I called Merc. Mag. they would nail it with just a SS rec. dc target voltage for the B+, but I'm also sure it would be a LOT more $$$$.


             Brad    
  

[OldHouseScott]

You could possibly get by with Doug's Fender reverb PT. 260VAC will give you about 360VDC after the FWB. The heater current with 2 6V6s and 3 preamp tubes will be about 2 amps leaving you with 1 amp for the relay supply. Hopefully Doug's relays are the low power variety. The B+ is only rated for 100mA, but you should be within that limit with the tube complement stated above.

[Willabe]

Yes that is an option and Doug's relays are low current draw.

But a PT wound for a FWFB will be smaller and lighter. On a PT with a CT on the B+ wind each half of the wind has to be able to support the PT's full B+ current. I also don't need a 5acv heater wind for a rectifier tube. On a Champ since it's class A the PSU is not going to sag very much because it's already loaded pretty much to the max. So no rectifier sag tone/feel effect. Plus with Power Scaling I can change the B+ just like swapping rectifier tubes. All these things will make the PT smaller, lighter because of less materials. 

I think 12.6acv heaters are nice because since it's twice the voltage it's half the current and so it's the lower current draw that decreases EM radiation from the heater wires. But to me more than that is with 12.6acv it's easier to wire up the 12_ _& tubes. Just pins 4/5 no crossing over the tube socket to pin 9. I like that.      To me that's the biggest pain of wiring up an amp.    

Also I can't use the dcv (70-80 dcv) stand off for the heaters and take my relay voltage from the same wind. (At least I don't think I can.)


                  Brad      

[OldHouseScott]

Not quite sure what you're saying, but Doug's reverb PT is designed for FWB operation -- no HV CT. You could still tie the heater CT to a standoff voltage and use the relay supply, just don't ground the negative terminal of the bridge. The relay coils don't care about the standoff voltage, although there may be some max rated breakdown voltage, coil to contact. As long as all the wiring stays inside the chassis, this will work, but it wouldn't be a good thing if you plan on using footswitches to activate the relays as you'd have the standoff voltage on the sleeve/shield line (and probably exposed metal plug covers). If you just don't like wiring up 6.3v heaters, well, I can't help you there   

[Willabe]

Not quite sure what you're saying, but Doug's reverb PT is designed for FWB operation -- no HV CT.

Hi Scott,

I forgot about that one and I already have 1 laying around on the shelf here.     

On a PT with a CT B+ secondary, lets say 250-0-250acv. So first the 1 side (top half wind in a schemo drawing) 250-0 (+) pushes voltage to 1 side (top half in a schemo drawing) of the FW rectifier. Then as the acv sine wave flips over, now the other side (bottom half wind in a schemo drawing) 0-250 (-) pushes voltage to the other side(bottom half in a schemo drawing) of the FW(HB) rectifier. So which ever part of the wind is pushing an acv to its rectifier, that part of the wind has to be able to support the full B+ dcvi]current[/i]. So it has double the wind needed that a Full Wave Full Bridge needs to supply the same amount of B+dc v and current. Because only 1/2 of the total CT B+ wind is being used at a time. (Fliping the PT's secondary B+ acv sign wave 120cps or 120hz.)

The old PT where made this way with the penalties of wasted materials, weight/size, chassis space and extra heat in the cab/head because it was cheaper in the long run with tube rectifiers. Also to do a FWFB with tube diodes you needed 4 tubes, 1 duel rec. and 2 singles, plus their extra heater current (IIRC it needed 3 seperate rec. heater winds) and space on the chassis. When reliable SS diodes became available the amp manufacturers would have had to re-do all their PT, they didn't. It made sense at the time but to me unless you need the voltage sag that a tube rectifier will give in some amps why use them? They cost more, can be less reliable, take more space on the chassis, need the extra heater wind and make the PT bigger/heavier. Only problem is where to get them?

That's why I called them for a bid. My understanding at this point is that once Edcor makes a PT and has it on file they will make it for anyone at a standard price not the 1'st 1 off build price. Now I very well could be wrong on this, but I never got that far with them untill I hit the ac/dc B+ snage with them.

You could still tie the heater CT to a standoff voltage and use the relay supply, just don't ground the negative terminal of the bridge. The relay coils don't care about the standoff voltage, although there may be some max rated breakdown voltage, coil to contact. As long as all the wiring stays inside the chassis, this will work, but it wouldn't be a good thing if you plan on using footswitches to activate the relays as you'd have the standoff voltage on the sleeve/shield line (and probably exposed metal plug covers).

Yep their for FTS's.

Still working on it, but here's the schemo for the build so far. (PT HT (B+) secondary is drawn with a CT which I hope to change. )


               Brad      

[sluckey]

On a PT with a CT....

Well, that's all true, but it's not a very good justification for an expensive custom built PT. Just get a nice Hammond that has a CT suitable for a 2 diode full wave conventional rectifier, 6.3 for all filaments, and use the 5vac to power the relays. And take the wife out for a movie and dinner with the savings.

[Willabe]

Well, that's all true, but it's not a very good justification for an expensive custom built PT.

That's the whole point. I was hoping that Edcore would be able to make 1 at a fair enough price the 1'st time and then if I bought a 2'nd, 3'rd, it would level off. Plus now it's on file at Edcore and anyone can get it at their off the shelf price? Still not sure about that part.

How do you figure the ac secondary voltage with a cap loaded FWFB? Is it secondary acv x 1.4?

I guess I could just give him a #, say 250acv and see what price he comes back with? At least I'd know if it's out of my reach.    

And take the wife out for a movie and dinner with the savings.

Wisdom speaks again.    

(I am cooking up a big pot of 15 bean soup with smoked ham hocks and a ham bone with the ham chopped off the bone to put in the soup, plus the smoked ham hock meat. Lousy restaurants around here and she does love my cooking. Gonna bake some corn bread to go with it.)


             Thanks,   Brad        

[sluckey]

Is it secondary acv x 1.4?

Yes. But that's for no load. Put a load on it and it gets trickier. Then you need to know how much load, AND, how much PT internal loss. 275vac @ 120ma might be more what you want for 360vdc.

BTW, what does FWFB mean? I know what you're talking about, but I've never seen FWFB before.

[PRR]

The transformer designer sells AC. Not DC. The gasoline station sells octane, not speed or range.

Especially with cap-input rectifier, the DC output is a sensitive complex function of winding resistances, capacitance, load resistance.

> smaller and lighter

Eh. If DC-only, perhaps 15% savings. When heater AC is also on the same core, under 10%. That's not enough to put you down to a smaller core-stamping. You might go down a few lams and a few ounces of copper.

> here's the schemo

NFB from VC won't work with 10uFd at the cathode.

I do not know what all the boxes in the power supply are. Not sure if there is sufficient ripple reduction going to the OT. SE stages have modest PSRR. We often do not want to feed them from the first stage of the cap-input rectifier. The least-weight/cost is another R-C filter stage. The R is a handy place to fine-trim the voltage to the final.

I'm sure the driver and preamp B+ decouplers could be larger than 1K.

240V, 200 ohms -- 44uFd will give 260-270V 100mA DC with about 12Vpp ripple.

[Willabe]

Yes. But that's for no load. Put a load on it and it gets trickier. Then you need to know how much load, AND, how much PT internal loss. 275vac @ 120ma might be more what you want for 360vdc.

OK, thanks Sluckey. I think I get it now.  

BTW, what does FWFB mean? I know what you're talking about, but I've never seen FWFB before.

I got in the habit of writing it that way from KOC's TUT books and the Power Scaling site. Full Wave Full Bridge, Full Wave Half Bridge. I guess he puts it that way for clarity? Maybe it's more confusing than clarifying because no one refers to it that way here?

(BTW, my wife loved the 15 bean soup and corn bread.    )    


             Thanks,   Brad      

[Willabe]

Eh. If DC-only, perhaps 15% savings. When heater AC is also on the same core, under 10%. That's not enough to put you down to a smaller core-stamping. You might go down a few lams and a few ounces of copper.

Ok, I thought it would be at least 25% to as much as 40%.

NFB from VC won't work with 10uFd at the cathode.

Doh!  The cap will short the OT ac signal to ground. So I have to get rid of the bypass cap or set it up like the driver in a Vibro Champ? Any suggestion on which would be a better way to go?

I do not know what all the boxes in the power supply are. Not sure if there is sufficient ripple reduction going to the OT. SE stages have modest PSRR.

The QV-LV (Quiet Supply-Low Voltage) is a mosfet capacitance multiplier/active hum filter <450vdc. IIRC Merlin has a version of one in his book on PSU? The VCK (Voltage Clamp Kit) is a mosfet based circuit that "acts like a varible zener diode" <600vdc. It keeps the B+ from rising as you unload the PT by turning down the PS/VVR. (It keeps the preamp stages B+ from rising and changing it's tone and helps kill the dead spot on the PS/VVR control.) The 2, SV84's are Power Scale circuits. One is for the power tubs plate/screen and the other is for the preamp tube and tone stack driver/recovery tube. It will have a low B+ voltage limit of ~100vdc.

I've wanted to try these circuits in an amp and thought this would be a build to try them on. The PCB boards will fit in easily and I also bought a few of Dougs new relay boards that I might use if I need them. I'm thinking about bridging the 2 input stages.

The B+ PSU will be > 44uF > QS-LV > VCK > 44uF for the plate supply then 1K for the screen. Drawing of PSU attached below.    

I'm sure the driver and preamp B+ decouplers could be larger than 1K.

Yes I just put that on the drawing, will increase as I build the amp. Maybe 2K2, 4K7?

240V, 200 ohms -- 44uFd will give 260-270V 100mA DC with about 12Vpp ripple.

You lost me on this.  Is this an example of a R loaded PSU? Or is the 200 ohms the PT HT winds resistance?


 
            Thanks,     Brad    

[jojokeo]

Brad,
With VVR you're going to be adjusting the voltages so no sense in getting too overly fine/concerned with the B+ voltage as they change when you turn the dial anyway. When I need to "guess" a B+ voltage on a scratch build using a FWB, I simply ÷ .79 to .76 to get close and depending on high or low load to be expected respectively. So a 280V PT will give you between 354V & 368V for your situation. This should be exactly what you're looking for? Then w/out VVR and you wanted as close as possible - simply adjust your resistor values.

[Willabe]

When I need to "guess" a B+ voltage on a scratch build using a FWB, I simply ÷ .79 to .76 to get close and depending on high or low load to be expected respectively.

Ahh, thank you.    

Then w/out VVR and you wanted as close as possible - simply adjust your resistor values.

 

The VCK has a voltage limit trim pot on it to set the B+ which is a nice feature.

With VVR you're going to be adjusting the voltages so no sense in getting too overly fine/concerned with the B+ voltage as they change when you turn the dial anyway. So a 280V PT will give you between 354V & 368V for your situation. This should be exactly what you're looking for?

Hmmm, maybe I need to re-think what I realy need for B+?

Fender AA764 Champ schemo says 320vac secondary, 360vdc B+ and Vibro Champ schemo has 315vac secondary, 355vdc B+ listed. Probably closer to 380vdc now a days with higher wall voltages? G. Weber in his books said he liked to use a 5V4 in a Champ so, 320x1.2=384? Now a days maybe 400vdc? He also said he actual liked a GZ34 better and only said he wrote about using a 5V4 because at the time he couldn't get a good GZ34. So now where at 320x1.3=416, maybe ~430vdc now a days.  

How much can a 6V6 take in SE/cathode bias? I know the tube doesn't see the full B+ in cathode as you have to subtract the cathode voltage.

So with all this to consider, Hmmm....

Maybe because the amp will have PS/VVR it would be best to leave the B+ on the higher side and use the PS/VVR to dial in the amps tone as far as a browner/darker/earlier breakup sound? Which would also leave me a higher B+ for the pre/tone stack tubes to play with since they will have their own PS/VVR circuit. It might be nice to go from say 270vdc for the plates like on a BF SR all the way down to ~100vdc. Power stage will have a master volume so I won't have to worry about over driving the power tubes grid.

Plus the reverb driver 12AT7 would probably like the higher B+ too. BF SR 12AT7 sees 450vdc.


              Thanks,   Brad      

[jojokeo]

I can see you haven't used a VVR before? You're planning and thinking waaaaaay tooooo much on this. The tone is not drastic at all and in fact is very very slight at best! The only thing you're going to get with a very high B+ is added noise and possible problems. Sorry, I don't want to burst your bubble but it is what it is.

[Willabe]

I can see you haven't used a VVR before?

No I haven't yet.   

The only thing you're going to get with a very high B+ is added noise and possible problems.

OK, so what would you consider very high B+ with a SE/cathode bias 6V6?

Sorry, I don't want to burst your bubble but it is what it is.

If/when I get goofie just use this on me,       or this,     



               Thanks,    Brad           

[jojokeo]

I don't think you need a very high B+ or even want it for only a SE 6v6 amp. You're only going to get that 5.5 watts of output power and the nuances, etc. aren't nearly the same as you get with much higher wattage amps. For a 5.5w amp the cone only moves so far. There's much more wiggle room and tolerances as it relates to "tone" on a low powered amp and the VVR or any other voltage change is not a big deal in and of itself. I'd say to stay w/ your 280v PT shooting for 360 B+ is perfectly fine, more than adequate. Gererally tubes running at the high-end borderline can get noisier, become brittle and harsh sounding, more prone to microphonic or osciallation, induced volatages, stray signal induction, become under biased, and load resistance changes on OT to name a few examples. Your main concern should be the lowering of the voltage on the plate/screen (via VVR) only while maintaining close to the same voltages on your preamp tubes for best results. Personally, VVR on a 5.5 watt amp seems hard to justify but for your first go at it, why the heck not?

[Willabe]

Ok, thanks Jojo.

The PS/VVR is for some friends who play in the worship band at church. If this don't get it then they might as well just play acoustic. Yes it should be a good build to get my feet wet with PS/VVR. If they don't like it I'll just keep it and use it at home for slide guitar without scaring my 2 dogs.    


            Brad      

[jojokeo]

What I've found is that VVR overdrives nicely at it's lower settings. When you dial down the power tube(s) and/or pi voltages the input sensitivity is greatly increased which helps with the early onset of overdrive and distortion. I think you'll like it and want it on a higher power amp where it's more practical.

[PRR]

> the PT HT winds resistance?

Yes.

> How much can a 6V6 take in SE/cathode bias?

12 Watts.

Or what jojokeo said: 5 Watts output.

Hardly matters if you run 200V 60mA or 400V 30mA, *except* you need to use a load impedance similar to your V/I ratio. And as plate goes up, screen should probably go down, an annoying complication.

Fender's method was "use available iron". A million parlor radios ran 6V6 at 250V with another 100V dropped in field coil. 350V when field coils went out of style. 12W/350= 34mA. 350V/34mA= 10K, but a 7K OT was available.