Power Supply School

ask your general tube related questions here

Power Supply School

Postby nineno » Tue Oct 26, 2004 10:07 pm

Well, since we've had so many good questions and answers about power supplies I think I'll jump into the fracas and ask some VERY rudimentary questions, but first some back ground…

I’m trying to build a junk-box tube amplifier. That alone is the motivation.

In essence I want the simplest possible amplifier with solid state rectification. Using these criteria I deduce that I’ll probably be building some 1.25Watt single-channel, single-ended “thing.”

At the base of this construction is a nearly-new power transformer that I have kicking around, a Hammond 269EX that is rated at 115V primary and 190-0-190 Vp-p. This is a remnant of a failed attempted at a headphone amplifier I tried to build earlier in the year.

Oddly, the transformer seems to put out much closer to 220Vp-p rather than 190Vp-p. I checked my mains voltage and it was 119.65Vac, so that explains a couple extra volts, but not 30.

The low voltage heater circuit also reads 7.4V, rather than 6.3. I even stuck a 6L6 tube on the heater secondary to load it down and see if it dropped (magically) into place…it did not.

I also stuck a simple center-tap rectifier on the circuit, and un-surprisingly the rectification was 440volts. If I loaded this circuit down with the normal HV amplifier load would the voltage fall in-line also?

Otherwise, what’s going on with my transformer? Did they forget a few primary windings or something?

Does anyone have any tips on building ‘the simplest/cheapest tube amplifier ever’ from the fore-mentioned transformer and any combination of the following tubes:

12AT7
12AU7
12AX7

5879
7199
6189
6077

6CG7

Thanks for the input guys, and thanks in advance for baring with me.

drew*
nineno
 
Posts: 117
Joined: Sat Aug 02, 2003 10:59 pm

Simplest ever!

Postby dhuebert » Wed Oct 27, 2004 7:19 am

Sorry! Didn't mean to get you excited with the subject line. I will be building a version of the Lighthouse-Electric SE 6BM8 amp from scratch(no kit). I dont see how you can get much simpler that one tube per channel. I know you don't have 6BM8 on your list, but I have more than enough twocents' for everybody.

Don
User avatar
dhuebert
KT88
 
Posts: 820
Joined: Thu May 01, 2003 9:26 am
Location: Winnipeg Manitoba Canada

Postby DortoH » Wed Oct 27, 2004 8:09 am

I am working on a SET using 6V6's or EL34's too. Hoping to go all Octal to ease the load on my fat fingers :P A 12AX7 and a pair of 6V6 would make Fender proud!

I have heard that Hammond's specs are a bit under-rated, although I would expect it to be a bit closer. That combined with the higher mains voltage may be the culprit, however I am far from an expert. You can contact Hammond too and they will let you know if that is within spec.

While I am on this subject, I have heard great things about the Hammond 125ESE SE output trannies and @ $35USD I thin I will be going with these. Apparently the published freq. response is very underated :) Someone measured them, and if I ever find the link I will post it, and they measured and listened real well.
DortoH
 
Posts: 100
Joined: Mon Aug 30, 2004 1:00 pm
Location: Great White North aka Canada

Postby erichayes » Wed Oct 27, 2004 11:27 am

Hi All,

Drew, I called my winder about this Hammond phenomenon a while back, and he said it is a combination of primary voltage and current capacity. He did a quick calculation (190/115 X 120 + 10% = 218V) and said they're going for current at the expense of physical size and heat generation. If you're going to be running your circuit conservatively, just adjust the resistor values in the filter section to reflect the higher voltage, and put low value resistors on the tube heaters a l? the ST35
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby DortoH » Wed Oct 27, 2004 12:38 pm

erichayes wrote:they're going for current at the expense of physical size and heat generation.


Based on that statement it sounds to me like if you are running it near or at capacity, the output should be as specified correct?
DortoH
 
Posts: 100
Joined: Mon Aug 30, 2004 1:00 pm
Location: Great White North aka Canada

Postby EWBrown » Wed Oct 27, 2004 1:28 pm

I had pretty good results with a 12AX7 / EL34 SE amp circuit that was very close to that shown by DortoH earlier. One "trick" I learned and tried, which made a noticeable difference, was to bypass the SE OPT B+ to ground with a 6.8 uF 630V Solen "Fastcap", this really brought out the higher frequencies, a real big "bang for the buck" mod/upgrade.

The 12AX7 has both sections paralleled, gives it a bit more gain and "overhead" capability. 12AT7 will also work, slightly less gain (70 vs 100 max) but doesn't make a lot of difference in overall power output.

Other similarly pinned out dual triodes should also work fine. Stay away from pentodes for the input stage, they just don't really cut the mustard.
If you need more gain, go an extra triode stage before the output tube.

The circuit I used is basically this one:

http://www.bonavolta.ch/hobby/en/audio/el34_4.htm

I made two variants of this amp, one uses One Electron UBT-3 OPTs in triode mode, the other uses James OPTs in UL mode, this gives more power output and still sounds great. Power tranny for both is Hammond 272HX (300-0-300 @ 200 mA) and solid state rectification (uF4007s) and LOTS of filter capacitor, 470 uF/450V for the B+s, ndividual 32+32 uF JJ caps for the 12AX7s. Cathode bias, 470 - 500 Ohms 5W resistor for each EL34. This gives about 56 mA bias, about 28VDC above ground.
No reason why Hammond 125ESEs shouldn't work fine with this design.

The NNETG folks gave this one high praise, it really rocked the house driving a pair of vintage ALtec A-7 VOT speakers.

These appear my "photo gallery"

/ed B in NH.

/ed B
Last edited by EWBrown on Thu Oct 28, 2004 11:15 am, edited 2 times in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby erichayes » Wed Oct 27, 2004 9:16 pm

Hi All,

That's it exactly, Rod. An ideal power transformer, with no IR losses, would put out its specified voltage(s) regardless of current, as long as the input voltage remained the same. Obviously, this is an unattainable goal.

What the transformer designer must determine when spec.ing is "How much can it cost?" and "How hot can it get?" With continuous duty transformers, where a constant load is being presented, those are fairly easy to answer. The generally accepted temperature rise is 40°C above ambient, which usually comes out to around 65°C (149°F). Knowing the losses of all the materials, and plugging them, along with the temp-rise figure, into a formula that I don't know, the designer is given the wire guage and core dimensions necessary--and how much it's gonna cost.

Power trannies used in dynamic applications, such as power amps running class AB are more difficult to size, because the load current is varying in direct proportion to the musical content if the amp is being played hard. This is where compromises have to be made, and they usually involve using smaller guage wire and less iron in the core. The result is that the regulation won't be as good, and the side effect is a hotter running transformer.

I spec my power transformers for a 40° rise at continuous output (in the case of the PP 6BQ5 stereo unit, 19 WPC @ 0.3% THD, both channels driven) for an indefinite time. This is done solely for better regulation. The pleasant thermal side effect is, at normal listening levels, the trannie runs at around 15° above ambient. This translates to roughly human body temperature if you have a mild fever.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby Shannon Parks » Thu Oct 28, 2004 6:42 am

erichayes wrote:This is done solely for better regulation. The pleasant thermal side effect is, at normal listening levels, the trannie runs at around 15° above ambient. This translates to roughly human body temperature if you have a mild fever.


This is fantastic, Eric.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington


Return to tube 101

Who is online

Users browsing this forum: No registered users and 8 guests