• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Help with linestage power supply

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,
i've been playing about with some old tubes i had lying around during the holidays (some old 6aq5's. Thanks to the guys here that helped me get a basic circuit:) ) and i've decided to go and do something decent. The sound of tubes was impressive, however that particular tube sucked. Alot. Very microphonic and i could hear what sounded like a spring reverb.....

Anyhoo.... i bought a transformer about a year ago and i was hoping i could get some help in using it for some circuits i've decided to build (don't try changing my mind on tubes, as they are bought already). Problem is i need to burn off alot of voltage, as it's a 350-0-350, and the line stage needs 250, and the phono 305 (i want to run both channels and stages off the one power supply. Don't get me wrong, i love dual mono, but the last one sent me poor). Any help would be much appreciated.

Here are the line stage i want to use:
 

Attachments

  • 6sn7_cf.gif
    6sn7_cf.gif
    7.1 KB · Views: 518

PRR

Member
Joined 2003
Paid Member
The idea is "Electronic Design", not "Electronic pot-luck".

> i bought a transformer .... Problem is i need to burn off alot of voltage, as it's a 350-0-350

350VAC * 1.414 = 500V DC. You can't easily buy 500V power capacitors. So already you are in trouble. Not to mention that a "350V" winding may be 385 or even 400V at light load. You can either buy special 500V caps, stack 300V caps, or add a Choke (and a bleeder to keep the voltage from soaring before the 6SN7 warms up).

You probably wanted a 200-250VAC winding, not a 350VAC winding.

Adapting that plan to work with 300, 400, even 500VDC is no big deal. Above 300V I would change the 47K resistor to 100K, maybe change the 820 to 1K. All the caps should be rated for the supply voltage, although actually you can probably stick with the 250V output cap because there is no way that cathode is going to get above 100 or 150V. (The 0.033uFd must be rated full no-load supply voltage: it will get that much at every switch-on.) But as long as you have excess voltage you should put some of it in the power supply filtering. 5K in the power line will drop 100V.
 
Thanks. I sort of started to work it out. I was having problems with psu designer, but lowering cap values sorted it out. I'm still having a bit of trouble working out the total current drawn on each of the stages, so if anyone can help, i'd appreciate it. It looks like about 5mA for the phono stage (both channels), but i'm not sure about the line stage. Around 15 for both?
 
Time out, please.

Dropping Volts in a resistor wastes energy as heat. IMO, there is an ELEGANT way to deal with the problem. Use choke I/P filtration. Neglecting losses, the 350 V. RMS from the power trafo comes out as 315 VDC. That's much, much, better!

Choke I/P filtration has additional advantages too. The O/P voltage is well regulated. As long as an appropriate bleeder resistor (10 KOhms for a 10 H. choke) is placed across the 1st filter cap., 450 WVDC filter caps. will not be over stressed. The available DC current is slightly greater than the AC RMS current; that's quite different from the 0.5 RMS available with cap. I/P filtration.
 
Tanstaafl, Eli. A choke input is a very good option with all the advantages you point out, but it has downsides, too. One biggie is the minimum current draw requirement- when you go below that, it stops acting as a choke-input filter and the voltage rises. You can:

hang a resistor across the output sized to guarantee minimum current draw (but there's the heat and inefficiency demon biting again), OR

depend on the timing of rectifiers and signal circuitry (good luck), OR

rate all components to withstand a high voltage (oops, there goes that advantage!).

There's no panacea for starting with bad engineering criteria, and trying to match a random transformer to a random circuit is a bad engineering criterion. PRR said it a bit more kindly, I think ;)
 
The idea is "Electronic Design", not "Electronic pot-luck".

There's no panacea for starting with bad engineering criteria, and trying to match a random transformer to a random circuit is a bad engineering criterion. PRR said it a bit more kindly, I think

Well, the 700VCT transformer, as it turns out, is the PERFECT transformer for your needs. Do what Eli suggested using a 10-15 henry choke and a 6X5GT rectifier and you are good to go.

John
 
The critical current, in mA., is V/L. It works out that 1K Ohm per Henry is the value of the bleeder resistor needed to ensure that the critical current gets drawn. 15 KOhms is the appropriate value of bleeder resistor for a 15 H. choke. 21 mA. is the critical current for 315 V. and a 15 H. choke.

As SY indicated previously, there are I^2R losses in the bleeder resistor. From that perspective, a large inductor, with its reduced critical current, is better. If the penalties in cost, size, and weight are acceptable, bigger is better.

BTW, a benefit of choke I/P filtration I neglected to mention previously is a cooler running power trafo when compared to cap. I/P filtration.
 
John, I figure about 25mA as a minimum current draw for a 15H choke. Is that correct?

For 300 volts 10Hy is the critical inductance with a bleeder drawing 2.5ma (10% of 25ma). It looks like Hamish is looking at ~35ma total for his stereo pre-amp, so about 7Hy is critical and 14Hy would be optimal. 10Hy would work fine if he made the beeder value smaller per Eli's suggestion.

Critical inductance (for a full-wave rectifier, choke-input supply, 60Hz mains) is calculated by dividing the total load resistance by 1130. Doubling it would give optimal inductance.

John
 
The minimum current for an ideal choke is given by:

Imin = 2*(sqrt2)Vrms/(6pi^2 * fL)

Plugging in 350VRMS, 15H, and 120 Hz, one gets 14mA. Generally, one adds 40-50% to that to account for the realities of cores (that's the basis of the approximate formula that Eli gives). 2.5mA will not cause this to act as an L-input filter even if the choke were ideal.
 
There is more than one way of looking at the design of a choke-input power supply. The common method I quoted (calculating critical inductance from total load resistance) is a convenient method made even simpler by some people by using 1000 as the constant rather than 1130. Doubling the inductance and only bleeding off 10-20% of the current allows one to use a power transformer with a lower current rating. Setting the bleeder resistor approximately equal to the load resistance halves the total load resistance thereby halving the value for critical inductance, so a smaller value choke can be used. One could probably get away with an input choke value only 50% greater than critical inductance in a class A circuit, as use of "optimal" inductance is simply conservative practice.

My source is Landee, Davis, and Albrecht's "Electronic Designers' Handbook" which has been a "bible" for me for many years.

John
 
Well, the simulation in PSUD indicates that for a 2.5mA current draw and a 6AX4 rectifier, 15H, 350-0-350, the output will be 424 volts. At 18mA draw, just over the minimum calculated by the formula I gave, the output is 312V. That voltage stays the same (accounting for DCR drops) at 25mA, which indicates that 18mA is over the minimum, 2.5 mA is under the minimum.
 
Thanks for the advice. I realise the transformer is probably far from optimal, but it was cheap. And some say it's too big, others not big enough.......the fact is it needs to be used, but next time i'll buy a smaller one.

When i first started playing around in psud, it kept killing the rectifier, but i slowly started working it out. Thanks for all your advice. I haven't had much experience with chokes, but i'm on a steep learning curve. It seems that with CLCLCRC combinations i can get the ripple to pretty acceptable (read VERY).

In regards to chokes, does the mains frequency play a part in selection? We have 50Hz here.....Back to the radiotron handbook.

Thanks:)
 
Well, the simulation in PSUD indicates that for a 2.5mA current draw and a 6AX4 rectifier, 15H, 350-0-350, the output will be 424 volts. At 18mA draw, just over the minimum calculated by the formula I gave, the output is 312V. That voltage stays the same (accounting for DCR drops) at 25mA, which indicates that 18mA is over the minimum, 2.5 mA is under the minimum.

If you read my post carefully, you will see that the 2.5ma drawn by the bleeder is in addition to the 25ma that is drawn by the circuit. For a load of ~300vdc at 2.5ma the critical inductance is 106Hy. The bleeder doesn't necessarily have to draw a lot of current for the power supply to have good regulation.

John
 
Ahh, you're right and I compounded it by just looking at the phono stage and thinking, "2.5 mA.". So, the minimum current is indeed satisfied after warmup.

It will still require higher voltage rated parts so that any one of a zillion small failures that could cause the tube not to conduct (especially in the heater circuit) will not cause a major chain reaction and big cap explosions.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.