Power transformer current, parallel SE 6V6 amp

[Diverted]

Hi all,

I have a single ended 20W 3K to 8 ohm output transformer laying around so I thought I would build a dual single ended 6V6 amp with it.

In a singe 6V6 version of this same amp design I just built I used a 275-0-275 secondary rated for 104ma with a few diodes to rectify. That gave me around 350 B+.

I can't find a transformer rated for that voltage at the right current. I am shooting for 160 to 200ma. I did find a Hammond transformer rated for 260V (no center tap) @173ma. Can anyone with better math skills than me figure out what kind of B+ voltage I'd be getting with that transformer and a full wave rectifier? Tube lineup is two 6V6s and three 12A types. Single channel champ circuit with Fender reverb, basically.

Thanks for any input on this transformer.

[shooter]

I used Edocr's XPWR050-120 on my last 3 6V6 amps, usually land somewhere around 340-350vdc
they have a small footprint if that matters, they also take 6 weeks to get

[sluckey]

In a singe 6V6 version of this same amp design I just built I used a 275-0-275 secondary rated for 104ma with a few diodes to rectify. That gave me around 350 B+.

That PT was just loafing with a single 6V6. I would use the same for a 2x6V6 amp.

[DummyLoad]

hammond 260G gets you close to what you specified. however, how hot are you going to idle those 6V6 to require a 200mA PT with 350V B+?


use ohms law. 6V6GT is good for 14W - 90% is 12.6W x 2 tubes is 25.2W total. ignoring V cathode if in auto bias, 25.2W/350V=72mA - so, why do you need a 200mA PT?


hammond 270EX is more than you need, 270DX will work.


--pete

[Diverted]

Not too small, current-wise?
Everything I've read on using paralleled 6V6s single ended suggests somewhere in the range of 140ma+ current rating on the secondary. I always thought that made sense from a layman's perspective given that a single 6v6 Fender Champ PT is rated at about 70ma.

[DummyLoad]

do the math & buy what you need. it's your money. 


--pete

[PRR]

6V6 at 12-14W Pdiss makes 5+W audio output. A 20W OT justifies *four* 6V6.

20W @ 3k is 343V 114mA peak. Taking plate-drop of like 50V this suggests 390V of B+ @ 140mA.

If you can only find two 6V6, you are looking at 10W-11W and probably 280V 100mA.

> a single 6v6 Fender Champ PT is rated at about 70ma.

That's a DeLuxe, loafing. The one-6V6 Champ runs 35mA-40mA plus a few for little tubes. It could use a smaller transformer but apparently Leo got such a good deal on DeLuxe PTs that he could use them in Champ also.

If you ran a more-sane 250V-300V supply, you could be aiming for more current (and a lower load impedance like 1.25k). But that wouldn't be Rock.

[Diverted]

Well that’s interesting about the Fender transformers. I know the champ and Deluxe used the same transformer but I always thought it was underpowered on the Deluxe  but about right for the champ. Thanks for that info.
I’m just going to use two 6v6s. Won’t have room for four.
Thanks!

[AmberB]

Just curious, have you ever thought about going with two 6L6s instead of 6V6s?  The output transformer will handle it...it would probably handle a 6550 or a KT 88 also...

[DummyLoad]

Not too small, current-wise?
Everything I've read on using paralleled 6V6s single ended suggests somewhere in the range of 140ma+ current rating on the secondary. I always thought that made sense from a layman's perspective given that a single 6v6 Fender Champ PT is rated at about 70ma.

restating PRRs reply:

looking at the 5C1 schema & layout, we see 18V at the cathode of the 6V6 across a 470R cathode resistor V/R=I so then 19V/470R= 40mA, that's with 340V B+. you'll be running 350V so to keep the anode dissipation as in the original, you'll need to idle at slightly less than 40mA - difference is not enough to warrant concern. with 2 6V6 you'd be pulling 80mA, with a perfectly matched pair.

i'd filter with a pi network before the plate supply & since you're using SS rectifiers, you have the option to use larger filters. if you go with the 5f1 values, and with 10" or larger speaker, buzz level may be intolerable to your ears. 

https://el34world.com/charts/Schematics/files/Fender/Fender_champ_5f1_layout.png

respectfully,


--pete

 

[Diverted]

Thanks Pete. I'm using the same schematic I used in a recent build, apart from doubling the 6V6s. There are a few other changes including rectifier, but this is mostly it. This layout has a an extra (I used 56UF) filtering stage. The amp is super quiet, actually. No hum though I did not use a choke. Thank you!

[nworbetan]

Not too small, current-wise?
Everything I've read on using paralleled 6V6s single ended suggests somewhere in the range of 140ma+ current rating on the secondary. I always thought that made sense from a layman's perspective given that a single 6v6 Fender Champ PT is rated at about 70ma.

restating PRRs reply:

looking at the 5C1 schema & layout, we see 18V at the cathode of the 6V6 across a 470R cathode resistor V/R=I so then 19V/470R= 40mA, that's with 340V B+. you'll be running 350V so to keep the anode dissipation as in the original, you'll need to idle at slightly less than 40mA - difference is not enough to warrant concern. with 2 6V6 you'd be pulling 80mA, with a perfectly matched pair.

i'd filter with a pi network before the plate supply & since you're using SS rectifiers, you have the option to use larger filters. if you go with the 5f1 values, and with 10" or larger speaker, buzz level may be intolerable to your ears. 

https://el34world.com/charts/Schematics/files/Fender/Fender_champ_5f1_layout.png

respectfully,


--pete

The power dissipated at idle isn't the same as the audio power output.  When your tubes are idling at 80mA for the pair, they're converting that 80mA into approx 26 watts of heat.  You'll want a power transformer that can provide at least twice the idle current.

The max theoretical efficiency of a class A tube amp is 50%, which means that in addition to every watt of power that goes into the speaker, the power tube anode(s) will dissipate at least that much (probably more) as heat.

There's a good example on the http://valvewizard.co.uk/se.html page with more realistic numbers than I'll be able to come up with.  :)

[PRR]

Welcome!

You'll want a power transformer that can provide at least twice the idle current.

Why?

Ideal class A would not change from idle to max. Most resistor-biased push-pull power amps have 5%-15% rise, but since we can't flog them to max ALL the time we don't have to account this too exactly. Fix-bias amps can have huge rise from idle to max: 3X for some tubes, 40X in some transistor amps.

[DummyLoad]

>> You'll want a power transformer that can provide at least twice the idle current.

really? why? cite references please. dc ratings for most tube amp full wave CT transformers are stated as being the actual DC load rating. class A output stage does not deviate much off idle. g2 does some at full output.

a 20W SE OT if designed correctly is designed with 40W power handling. the designers assume that in SE power to load will be less than 50% - typically, less than that, around 40%.

--pete

[PRR]

...a 20W SE OT if designed correctly is designed with 40W power handling. ...

The OT does not see 40 Watts of anything. At idle it drops less than 10% of the supply voltage, so 4W-5W of heat. At full roar it handles the 20W of AC Audio.

[DummyLoad]

...a 20W SE OT if designed correctly is designed with 40W power handling. ...

The OT does not see 40 Watts of anything. At idle it drops less than 10% of the supply voltage, so 4W-5W of heat. At full roar it handles the 20W of AC Audio.

thanks for the clarification.


--pete

[shooter]

At idle it drops less than 10% of the supply voltage

...years later a bulb lights
measured that drop many times, then just "lumped" it into the tube and called it 20W, from now on the tube only gets 16W!
thanks also for flippin a switch.


I have always used the 20W OT "spec" to decide AC, i want the OT to just pass it along without much effort so I typically go one size up.
that stops at 25W Audio in SE for weight n cost reasons

[PRR]

> bulb lights measured that drop many times, then just "lumped" it into the tube and called it 20W, from now on the tube only gets 16W!

Wiring error. The transformer losses do not work in our favor. For 20 Watts sine OUTput, the TUBE idle heat is at-least 40W, and in practice like 50 Watts (40% efficiency). To that we must add OT and bias-resistor losses. For good economy (to buy and operate) the OT losses will be 5%-10%. For most self-bias tubes the cathode resistor loss is 5%-10%. So we are looking at 54W of heat for 20W of sine output. (Before heater power....)

[shooter]

Wiring error

English (language) error 
I just never thought "through" the series Vdrops at idle like you expressed, they just "were".
the AC dissipation I have down, just not verbally well   

[nworbetan]

Welcome!

You'll want a power transformer that can provide at least twice the idle current.

Why?

>> You'll want a power transformer that can provide at least twice the idle current.

really? why? cite references please.

Thanks for pointing out my mistake.

When I originally believed what I said, I almost qualified it with "If you want to minimize sag and intermodulation distortion with power supply ripple, you'll want a..."  But now that I've thought about it a bit more I realized the reservoir cap attached to the plate is a source of current parallel to the PT for the part of the output cycle when the control grid approaches zero and the plate current is at its highest.

So instead of expecting the PT to handle all of the peak plate current, the way to minimize sag and IM distortion is to use a reservoir cap that's big enough.  That being said, sometimes minimizing those things is the wrong thing to do in a guitar amp and I'll leave it to your ears to be the judge of whether that's what you actually want or not.

[PRR]

The PT current rating is a matter of long-term heat. Any transformer can supply much more current short-term.

The PT only supplies current 100/120 times a second. Most musical frequencies are higher than this. The PT can't cover the musical peaks.

As you say, the value of the cap is a matter of taste.