Author Topic: 0-350V 0.2A Bench Power Supply  (Read 9222 times)

001 and 1 Guest are viewing this topic.

Offline Ysjoelfir

  • Regular Contributor
  • *
  • Posts: 214
  • Country: de
  • Collecting vintage Stuff and throwing stuff around
    • Tinkertubes Lab
Re: 0-350V 0.2A Bench Power Supply
« Reply #75 on: November 22, 2017, 06:40:03 AM »
V1 and 2 are just voltage stabilizing tube with a burning voltage of 150. You could substitute them with suitable zener diodes, potentially some of them in series.
V3 is the tube that drives the series pass element. You would have to calculate the grid voltage that needs to be present there to get the original controll range (meaning to controll the 6L6 from "flood that glass bulb with electrons" to "keep those electrons around that glowing stick") and build a control stage corresponding to that value. This could be done in many ways, pick one you feel comfortable with and start calculating :)
« Last Edit: November 22, 2017, 06:42:37 AM by Ysjoelfir »
Greetings, Kai \ Ysjoelfir
Ysjoelfir @ Tinkertubes Lab
 

Offline oldway

  • Super Contributor
  • ***
  • Posts: 1817
Re: 0-350V 0.2A Bench Power Supply
« Reply #76 on: November 22, 2017, 10:32:31 PM »
The IP17 Heathkit power supply does not meet your specifications: nominal 100mA, 125mA max instead of 200mA.
If you have bought one, I advise you not to try to modify or to improve it, except maybe replace V1 and V2 (OA2) with zener diodes.
I have never been a fan of these cold cathode regulator tubes, but it was the only thing available at the time.
http://www.sgitheach.org.uk/ip17.html

If you want to design and manufacture your own high voltage power supply, you must follow this procedure:

1) The biggest problem is the transformers.
Can not make a  "custom" transformer for a single unit, the price would be excessive.
It is therefore necessary to use commercial transformers.

So the choice is limited and it will be necessary to use several transformers.

One option is to find out if a transformer for a tube amplifier could serve ... They usually have at least one high voltage winding and a 6.3V winding.
https://tubedepot.com/t/diy-central/transformers-and-chokes?page=1

Another option is to use for high voltage an isolation transformer that has a 400V winding. (using secundary as primary, no problem, transformer is reversible)
https://www.erea.be/fr/produits/detail/efsp160
The rectified high voltage (bridge rectifier) would then be 565V, which allows regulated output voltage up to 450V, which is ideal for those working on vacuum tube projects.

To be continued....
« Last Edit: November 22, 2017, 10:35:10 PM by oldway »
 

Offline b_force

  • Frequent Contributor
  • **
  • Posts: 757
  • Country: 00
    • One World Concepts
Re: 0-350V 0.2A Bench Power Supply
« Reply #77 on: November 22, 2017, 10:41:17 PM »
You can also use two transformers back-to-back.
But there are plenty of affordable HV transformers out there.

"If you can't explain it simply (or at all), you don't understand it well enough." A. Einstein

http://www.oneworldconcepts.com/
 

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #78 on: November 22, 2017, 10:49:07 PM »
Hi

Thanx for yours opinion


But read post abowe carefully. I found awesome industrial grade siemens transformer. Awesome 230/400v transformers for decent price You can find at undeground and ubach reilroad workshops
Some of this workshops can rewind it for You. So it is not problem

I see schematic of ip17 is very basic and simple. I want to substitute el34 with pl509 (deathproof beast for decades of tube tv, 500ma catode current, $3 at garage sales) . Can i eliminate negative supply if it is not critical to have 0 zero minimum? Can i use pass tube in triode mode to elimenate screen power supply?
« Last Edit: November 22, 2017, 11:39:19 PM by 001 »
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #79 on: November 22, 2017, 11:50:34 PM »
IMHO that tube would do nicely in a hybrid supply - IC current and voltage control loops, discrete transistors for the grid driver and a tube for the main pass element.

http://www.radiovilag.hu/images/PL509.pdf

Its got a 40V heater.   If you capacitively couple a bridge rectifier to the heater supply, its trivial to derive a boosted screen grid supply 50V above the anode voltage.   Use a 20-0-20 transformer, connect the center tap to the cathode (as the max heater-cathide voltage is only 250V so you cant run the heater grounded) and add another bridge rectifier you can also use it for a floating +/- 15V regulated supply riding on the output voltage for OPAMPs in the control circuits.  The PL509 needs at least -50V grid bias for cutoff, so you'll need another capacitively coupled rectifier to get that from the filament supply. 


However the fly in the ointment is the max 30W plate dissipation.   At 400V in, that means you cant pass more than 75mA without overheating the plate with a risk of catastrophic envelope failure if sustained.   Either you are going to need 3x PL509 in parallel or you are going to need a tap-changer that tracks the output to keep the average Vak below 150V for 200mA output.

One approach  would be a 55-0-55, 50VA transformer and a 110-0-110, 100VA transformer, which after rectification gives you voltage steps of 77V and 155V respectively.  The easiest option for tap changer control would be a MCU monitoring the output driving relays.     There's enough headroom there to extend the spec to 0-400V, 0-250mA out.
« Last Edit: November 22, 2017, 11:59:06 PM by Ian.M »
 

Offline richard.cs

  • Frequent Contributor
  • **
  • Posts: 511
  • Country: gb
  • Electronics engineer from Southampton, UK.
    • Random stuff I've built (mostly non-electronic and fairly dated).
Re: 0-350V 0.2A Bench Power Supply
« Reply #80 on: November 23, 2017, 12:05:44 AM »
However the fly in the ointment is the max 30W plate dissipation.   At 400V in, that means you cant pass more than 75mA without overheating the plate with a risk of catastrophic envelope failure if sustained.   Either you are going to need 3x PL509 in parallel or you are going to need a tap-changer that tracks the output to keep the average Vak below 150V.

Note that that is with the design centre rating system*, not an absolute maximum which is given as 40 W. Switching to a PL519 would get you an extra 5 W anode dissipation on an otherwise very similar valve: http://www.r-type.org/pdfs/pl519.pdf - it's essentially an uprated version.

*i.e. the manufacturer means you can design for 30 W at nominal mains voltage and component values (10% resistors, etc.) and it won't blow up in real world use. The switch from design centre to absolute maximum was a big part of why early transistors got a reputation for being fragile and easily damaged, coupled with the lower heat capacity of a microscopic silicon die compared to an anode or grid.



 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #81 on: November 23, 2017, 12:31:41 AM »
400V DC bus in, 0.2A cathode current is going to be 80W plate dissipation so even with a PL519 you'd need two in parallel to operate without a tap changer, and they would be running *HOT* at full load and low output voltage.   

Unless the tube in question is popular with the tube amp audiophiles, long term availability is likely to be an issue, so personally I'd rather minimise the number of 'bottles' required so I could afford to stock enough spares.  Keep it down to 30W and enjoy long tube life and be able to use either PL509 or PL519, depending on cost and availability.
« Last Edit: November 23, 2017, 12:39:06 AM by Ian.M »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 5471
  • Country: us
  • DavidH
Re: 0-350V 0.2A Bench Power Supply
« Reply #82 on: November 23, 2017, 12:54:02 AM »
Jim Williams used an Eimac 75TH with a plate dissipation of 75 watts in his example shown below.

Foldback current limiting would be a good idea in a general purpose power supply to limit dissipation at low output voltages.  A time constant can be used to allow maximum current for short periods.
 

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #83 on: November 23, 2017, 02:54:03 AM »
Simple circuits attached
Is it usable? Comment please
« Last Edit: November 29, 2017, 01:32:24 AM by 001 »
 

Offline Hero999

  • Super Contributor
  • ***
  • Posts: 9002
  • Country: gb
Re: 0-350V 0.2A Bench Power Supply
« Reply #84 on: November 23, 2017, 03:08:46 AM »
Jim Williams used an Eimac 75TH with a plate dissipation of 75 watts in his example shown below.
I posted the same circuit awhile ago, except it used a BJT, rather than a valve. The main issue is there will be a significant minimum load requirement, due to the zener current and using 1k2 for the upper resistor in the LM317's potential divider. Reducing the resistor values and adding a PNP transistor, as in my circuit, would eliminate the need for the minimum load, at the expense of a higher quiescent current.

http://www.eevblog.com/forum/beginners/design-of-circuit-for-0-24v-5amp-regulated-power-supply/?action=dlattach;attach=352281;image

EDIT:
The above circuit won't work up to 400V. It will exceed the safe operating area of most BJTs, unless many devices are connected in parallel. Use a MOSFET, which can tolerate >0.2A @ >350V, such as the FQL40N50.


http://www.mouser.com/ds/2/308/FQL40N50-1124413.pdf
« Last Edit: November 27, 2017, 10:30:45 AM by Hero999 »
 
The following users thanked this post: 001

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #85 on: November 23, 2017, 03:22:08 AM »
@001,
The first circuit is unregulated, and as the IRF740 doesn't have a DC line on its S.O.A graph, it probably wont survive a sustained output short to ground even if it survives the initial transient.

The second circuit is regulated, but doesn't have a wide output voltage adjustment range, nor can it easily be modified to extend its adjustment range below the voltage stabilizer tube's operating voltage.  Also the only current limit is the pass valve cathode emission limit.

@Hero999 & David Hess,

I believe that Jim Williams circuit has severe issues if lightly loaded or if the output is ever shorted.  Look at the Eimac 75TH datasheet for the grid cutoff voltage with 2.5KV on the plate and you'll see why I'm concerned.
 
The following users thanked this post: 001

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #86 on: November 23, 2017, 03:55:30 AM »
@001,
The first circuit is unregulated, and as the IRF740 doesn't have a DC line on its S.O.A graph, it probably wont survive a sustained output short to ground even if it survives the initial transient.

The second circuit is regulated, but doesn't have a wide output voltage adjustment range, nor can it easily be modified to extend its adjustment range below the voltage stabilizer tube's operating voltage.  Also the only current limit is the pass valve cathode emission limit.

Thanx a lot!

Can You Help me combine theese two circuits?
I want use simple 0.2A current limitter from first circuit and tube pass from second circuit
but how I can substitute small pentode (actualy tetrode) driver to transistor (or simple common triode)?  :-//
« Last Edit: November 23, 2017, 03:57:28 AM by 001 »
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 9331
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: 0-350V 0.2A Bench Power Supply
« Reply #87 on: November 23, 2017, 04:04:22 AM »
@001,
The first circuit is unregulated, and as the IRF740 doesn't have a DC line on its S.O.A graph, it probably wont survive a sustained output short to ground even if it survives the initial transient.

Old IRF740 (IR) datasheet shows thermally limited DC SOA.

I tested a new production IRF740 (Siliconix), at 300V, at significantly better than Rth(JC) max.  You get a lot of silicon for not much cost.

Current Siliconix datasheet probably doesn't specify DC SOA because of laziness or CYA.

It's not enough power in a single device for present needs, but about three in parallel would handle it.

001,

Are you looking to learn about power supplies, or are you looking for a finished, proven schematic that you can build?  There has been quite a lot of discussion in this thread about the former, and I'm beginning to suspect you want the latter.

Tim
Seven Transistor Labs, LLC
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
 

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #88 on: November 23, 2017, 04:11:33 AM »
Yes
I preffer complete shematic
but now I see what hybrid variant (pass tube and transistor driver) will be ok for me
how to combine that?
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #89 on: November 23, 2017, 04:34:03 AM »
Can You Help me combine theese two circuits?
I want use simple 0.2A current limitter from first circuit and tube pass from second circuit
but how I can substitute small pentode (actualy tetrode) driver to transistor (or simple common triode)?  :-//
Its not possible to combine them due the the difference in gate/grid control action between an enhancement MOSFET, and a valve, JFET or depletion MOSFET.

However if you start with the second circuit:

The 6Zh1P cross references to 6AK5

Delete everything to the right of and below the 6AK5  apart from its 10K-3K9 screen grid divider, and swap out the 1Meg anode feed resistor for a 100K resistor, then you could put a TL431 shunt regulator (with a potential divider off the output for its Adj pin) in cascode with the 6AK5, with its control grid fed off a 12V Zener (with a pullup to unreg +HT), and also use a NPN transistor pulling down the cascode cathode and a low side current sense resistor to implement current limiting as discussed earlier in this thread.  However the  6AK5 is only rated for 180V Vak, so it wouldn't be suitable for more than a 200V PSU.

Use the right valves and the basic idea would be good for a much higher voltage,  and with a TL431 providing the regulation, you could get the output down to a couple of volts minimum output.  However without a tap changer or switching preregulator, unless you use a brute of a transmitter valve as the pass element, you would run into plate dissipation issues with a high input voltage if its run for long in its current limiting mode.
« Last Edit: November 23, 2017, 04:48:13 AM by Ian.M »
 
The following users thanked this post: 001

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #90 on: November 23, 2017, 04:47:56 AM »
Awesome
Thanx  :-+

Can You coment theese ideas?

« Last Edit: November 23, 2017, 05:19:22 AM by 001 »
 

Offline oldway

  • Super Contributor
  • ***
  • Posts: 1817
Re: 0-350V 0.2A Bench Power Supply
« Reply #91 on: November 23, 2017, 04:48:59 AM »
With a winding of 400V and a rectified voltage of 565V max, we will have a dissipation of about 100W in the power tubes at 0V 200mA.

If we have a winding 230V, or 325V rectified, we can do 2 ranges: one from 0 to 320V and one from 280 to 450V.
The anodic dissipation at 200mA will then be: range 1, 60W max, range 2, 40W max, which is quite reasonable and even allows to envisage a current greater than 200 mA.

I reject the custom transformer solution because it is not applicable for many.

The other transformers:

For 6.3V heating of the power tubes, a 6V transformer can be used.

For the supply of G2 of power tubes (I will explain later why it MUST be pentode, and not triode) , it is possible to use a 6V / 110V + 110V transformer powered by the 6V winding of the power tubes heating transformer.
One of the 110V secondary will be used for the G2 supply, the other to generate the negative bias voltage if the isolation level between the 110V windings is sufficient.

For the heating of the other tubes, another small 6V transformer

All these transformers are easily found.

To be continued....

« Last Edit: November 23, 2017, 04:51:27 AM by oldway »
 
The following users thanked this post: 001

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #92 on: November 23, 2017, 04:58:47 AM »
UGH!!! More HV PSUs from the dark ages.  The second one is at least somewhat elegant with a long tailed pair for the error amp.  The first/third is just fugly.
 

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #93 on: November 23, 2017, 05:23:08 AM »
UGH!!! More HV PSUs from the dark ages.  The second one is at least somewhat elegant with a long tailed pair for the error amp

Special for You!
Can You comment it?
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #94 on: November 23, 2017, 05:50:50 AM »
Well its discrete silicon, Russian and as the KT940A is a 300V 0.1A 10W device it doesn't have a snowflake's chance in hell of meeting your specs or surviving an output short-circuit.
Please don't waste out time with small variations of the same concept that you haven't even checked against your specified requirements.
 

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #95 on: November 23, 2017, 06:13:11 AM »
it doesn't have a snowflake's chance in hell of meeting your specs or surviving an output short-circuit.

Is tube pass circuit will survive?
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 5750
Re: 0-350V 0.2A Bench Power Supply
« Reply #96 on: November 23, 2017, 06:17:40 AM »
Yes.  A 30W anode dissipation valve has a *FAR* higher transient overload capability than a 100W Silicon transistor.   As long as you have some sort of current limiting circuit to 'throttle back' or trip before the anode or envelope melts, it will survive gross abuse.
« Last Edit: November 23, 2017, 06:22:36 AM by Ian.M »
 

Offline Hero999

  • Super Contributor
  • ***
  • Posts: 9002
  • Country: gb
« Last Edit: November 23, 2017, 09:47:13 AM by Hero999 »
 

Online 001

  • Regular Contributor
  • *
  • Posts: 208
  • Country: 00
Re: 0-350V 0.2A Bench Power Supply
« Reply #98 on: November 23, 2017, 06:37:56 AM »
Why not combine the design I posted, using LM317 & the BJT cascode, with the one posted by David, using the valve?

how to do it?
 

Offline oldway

  • Super Contributor
  • ***
  • Posts: 1817
Re: 0-350V 0.2A Bench Power Supply
« Reply #99 on: November 23, 2017, 07:14:34 AM »
2) choice of power tubes.

The choice must be made according to the availability of electronic tubes.
Only tubes with 6.3V heating are still manufactured.
For European tubes, they were tubes whose name begins with E.
(the P meant 300mA series heating, the voltage varies from one type to another)
We therefore remove all tubes starting with P.

I heard mentioning tubes like PL500, PL504, PL509 and PL519.
They were tubes designed to operate with very high voltages such as those found in the black and white TV flyback stage (PL500 and PL504) and TV's colors (PL509 and PL519).
These voltages were several kilovolts, which is why the anode connection was at the top of the tube and not the socket to avoid flashbacks.

In an HV power supply, one only finds about 600V, which is a little bit for a tube.

The main suitable tubes are: EL34, KT88, 6v6GC

Considering the prices and the number of manufacturers, it is the EL34 that I selected.

No problem with its ability to withstand 600V (I knew a public address amplifier using 4 EL34 push pull operating with an anode voltage of 700V)
The anode power is 20W, so it will take 3 in parallel.
To equalize the currents, it is necessary to put a resistance in series with the cathode of 47R 1W.

Choose steatite octal sockets of good quality.

The heating current will be 3 x 1.5A = 4.5A for 6.3V. (6V acceptable)

To be continued......
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf