help needed on hi-V power supply circuit

A little over a year ago, I posted a request for help (1) in designing a circuit to get 300VDC out of a power supply that was pumping out ~520VDC. I got many helpful replies, but unfortunately, I was unable to complete the circuit. I now have three transmitters (2,3,4) that would like a nice source of high voltage and I must now admit that I really do need your help in completing this project.

I own a really nice, heavy transformer that I'd like to use- unfortunately, the voltage is a bit too high- it puts out ~520VDC (after a 3-pole filter 10uF/10H/10uF with 40kOhm bleeder resistor). I am looking for some ways to bring this down to the voltage I need.

Although I consider myself a competent builder, I am unfortunately not much of a designer - this is what has prevented my progress for the past year! I own the '05 ARRL handbook and Art of Electronics, but I'm just not comfortable doing this myself. Specifically, telling me the type of transistor/tube and not specifying the part number or an article describing a similar HV regulator would leave me somewhat stranded. I know I'm asking a lot here -- please understand that I am immensely greatful for any of your help!

Here is my specification / wish-list, in decreasing order of importance:

  • produce ~370Vdc at 120 mA
  • cost less than
  • not require modification of the transformer
  • also be able to produce ~240Vdc at 40 mA
  • also be able to produce ~300Vdc at 70 mA
  • be hollow-state
  • be elegant

References (1) One link to this thread is at

formatting link
(2) The One-Tube Three-Band Novice Transmitter: see
formatting link
and download the article on the mid-bottom right side of the page. (3) The Novice Special transmitter: ARRL members only, download at
formatting link
(4) The Simple Transmitter: ARRL members only, download at
formatting link

Reply to
Ross Tucker, NS7F
Loading thread data ...

Have you considered a choke input filter, instead of a capacitor input filter? The capacitor input filter approaches 1.414 times the AC RMS voltage at no load. The choke input filter approaches .9 times AC RMS voltage at the minimum load that keeps the current continuous through the choke. That difference in ratio means that the output will drop to something like 64% of what comes out of the capacitor input filter, reducing your ~520 volts to ~330 volts. It allows a higher DC average current from the same transformer without overheating it, too.

The minimum current load on the filter that will keep the current continuous and thus force the output to be the average of the input is about 30 mA (using your 10H inductor and 20uF after it). With load current less than this, the voltage will rise toward the same ~520 volts you get with the cap input filter as the load approaches zero.

formatting link

Reply to
John Popelish

From your description, I think you have this:

---+----[10H]----+-------+---> +V |+ |+ | [10uF] [10uF] [40K] | | | ---+-------------+-------+---> Gnd

If that's what you have, change it to this:

---[10H]---+------+------+---> +V |+ |+ | [10uF] [10uF] [40K] | | | -----------+------+------+---> Gnd

Measure the voltage from +V to gnd.

Ed

formatting link

Reply to
ehsjr

As a first, simple, experiment disconnect the 10uF nearest the rectifier and reconnect it in parallel with the other 10uF. Then see what DC voltages you get when drawing currents between 120 and 230mA. If it is near enough to the

370v you require, then add suitable resistors to deliver the other supply voltages and currents you've specified and you've got what you need.
Reply to
John A

formatting link

With *no* re-winding of the transformer:

1) Use a VR150 in series to drop the 520V to 370V 2) Use a 35V "control" transformer secondary in series opposing to the primary of that power transformer. 3) Get another power transformer.

ewinding a power transformer is no big deal.

Reply to
Robert Baer

formatting link

I did not mention that, evev tho i thought of that befiue the other solutions. Isn't it rather strange that it is well known that a choke input filter not only reduces the output voltage, but also gives better regulation - and almost never used?

Reply to
Robert Baer

Let's start at the beginning.

You have a power transformer and a rectifier, followed by a C-L-C filter (aka capacitor input). And you're getting a bit over 500 VDC from it.

1) What rectifier circuit are you using?

a) full-wave center-tap b) full-wave bridge c) voltage doubler d) some other type (specify)

2) What sort of rectifier(s) are you using?

a) High vacuum diodes (5U4G, 5Y3GT, 5R4GY, etc.) b) Solidstate silicon diodes (1N4007, etc.) c) some other type (specify)

3) What are the transformer ratings, part number, etc.?

4) Which of the listed transmitters have you built?

73 de Jim, N2EY
Reply to
N2EY

Capacitors are much cheaper.

Cheers! Rich

Reply to
Rich Grise

I'm not an radiomator nor an english language speacker. But I'm an experienced engineer, designing various things in the past 25 years of activity. One of these was a programable

30V-300V DC linear power supply with 1 V output resolution, 150mA output current and less than 10mV output ripple.

Your input voltage is too big for this purpose and three about your issues can't be obtained if you really want a rugged solution:

- no require transformer modification

- being elegant

- be hollow state (I'm not very sure what that exactly means)

The ideea with power resistor dividers where used back in the '75 (at least in the Eastern Europe) so I've guess in the USA back in the '60. You realize is far away to be an elegant solution.

Even the linear power supply is not elegant even most of radiomator equipments still have tubes and needs a hole basement (and an angry wife) to be functionaly. I suggest a switching supply (more or less similar to the TV HV stages) or if you don't want that dropping a some wires from your transformer secondary (also a series capacitor with your primary will do the same).

best wishes, Vasile

Reply to
vsurducan

Ross,

Try to find a used Heathkit IP-17 power supply on ebay or copy portions of the circuit diagram you think you can use.

formatting link

cheers, skipp

formatting link

: A little over a year ago, I posted a request for help (1) in designing : a circuit to get 300VDC out of a power supply that was pumping out : ~520VDC. I got many helpful replies, but unfortunately, I was unable : to complete the circuit. I now have three transmitters (2,3,4) that : would like : a nice source of high voltage and I must now admit that I really do : need your help in completing this project.

: I own a really nice, heavy transformer that I'd like to use- : unfortunately, the voltage is a bit too high- it puts out ~520VDC : (after a 3-pole filter 10uF/10H/10uF with 40kOhm bleeder : resistor). I am looking for some ways to bring this down to the : voltage I need.

: Although I consider myself a competent builder, I am unfortunately not : much of a designer - this is what has prevented my progress for the : past year! I own the '05 ARRL handbook and Art of Electronics, but I'm : just not comfortable doing this myself. Specifically, telling me the : type of transistor/tube and not specifying the part number or an : article describing a similar HV regulator would leave me somewhat : stranded. I know I'm asking a lot here -- please understand that I : am immensely greatful for any of your help!

: Here is my specification / wish-list, in decreasing order of : importance: : * produce ~370Vdc at 120 mA : * cost less than $50 : * not require modification of the transformer : * also be able to produce ~240Vdc at 40 mA : * also be able to produce ~300Vdc at 70 mA : * be hollow-state : * be elegant

: References : (1) One link to this thread is at :

formatting link
: (2) The One-Tube Three-Band Novice Transmitter: see :
formatting link
and download the : article on the mid-bottom right side of the page. : (3) The Novice Special transmitter: ARRL members only, download at :
formatting link
: (4) The Simple Transmitter: ARRL members only, download at :
formatting link

Reply to
Skipp says hello

Yes, John Popelish and Robert Baer suggested that as well. Choke- input filter circuits are nice and quiet, but they do suffer from voltage sag under load. If Ross measures his DC output voltage with the choke- input configuration, both with light load and full load, perhaps we can suggest a simple circuit to stabilize the output voltage at 300 volts. Dear me, I do feel a few ACSCII drawings coming on. In the meantime, I'm going to brush up on the irfpc40 600-volt power MOSFET I suggested last year.

formatting link
formatting link
?Ref=210355&Row=711822

Reply to
Winfield

That big old Triode? .... again.

A G C | . / 520V---+----| . | | | . \\ | | +-----+---------->300V | | | | +----/\\/\\---+ | | 1meg | 75k \\ | \\ 5W / | /200k \\ | \\ | \\| | | npn |-----+---+ |

Reply to
Tony Williams

Hmm... that datasheet has this alarming "scanned" look to it, so I decided to look up what IRF has to say about the part. The IRFPC40 is flagged as "divested", and when you click on "divested", you get redirected to some Vishay page. I guess it means "obsolete". So I went to to IRF's parametric search for a modern replacement, and to my surprise they don't list any 600V FETs at all -- they go only to 300V. What the... Didn't IRF use to be a big name in power MOSFETs?

Infineon has a broad line of HV parts, if you don't mind that 90% of each page load time is taken up by various photos of a yuppie playing with a laptop on a glass table. It sure bothers me.

--
robert
Reply to
Robert Latest

Yes, but last month IR sold their entire MOSFET line to Vishay. I should have given the new Vishay datasheet link,

formatting link
It's just the same old IR scanned document, with Vishay's name added to the bottom of each page, and an acquisition announcement on the last- page. I guess along with their Siliconix MOSFET product line, Vishay is going to compete for the new powerhouse title.

I see DigiKey has substantially raised the price of the IRFPC40, and since it was only a 150-watt MOSFET anyway, I'm sure we can find a better choice. Maybe Fairchild's 208-watt TO-247 600-volt FCH20n60, $3.58 qty 10 at DigiKey.

formatting link
formatting link
?Ref=29770&Row=612787&Site=US

Or maybe Fairchild's 240-watt SPW24n60C3, at $4.13 or what the heck, go for it, Fairchild's FCH47n60 at $9.98 each, single pieces. The 47n60's

417-watt spec (calculated from theta-JC = 0.24C/W) will allow a little higher current rating for the regulator.
formatting link
Reply to
Winfield

Yes, I remember that one. The nice thing about a tube is the way the transistor's load resistor is bootstrapped by the cathode, nicely starting and setting the operating current of the NPN, and the zener if you lose the 75k 5W (it's not needed is it?). Let's replace the triode with a Supertex depletion-mode power MOSFET, to keep the nice self-biasing feature. The DN2540N5 is a 400V 150mA part in a TO-220 case, and Mouser has 1704 of them at $1.65 each, $1.33 qty 100. We'll need to cascode the depletion-mode MOSFET with a regular power MOSFET for more power-dissipating capability, and for a higher voltage rating (to handle a short). We can also add a short-circuit current limit, which I've shown in a simplified form; with three more resistors it'll be a more suitable foldback type. The NPN is running at 0.25mA,

50mW. 470k, 1/2 W 68k ,--/\\/\\--+--+---+-----/\\/\\-----, | | | | | FCH47n60 ___| | '-- C E -----+ | ------ Z B | | | | | | | --+-----' '--+--, ,---+-/\\/\\-+-+---+-->300V 520V _|__|_ | | DN2540 ---- | min |____ZA_____| min load | | resistance +----/\\/\\---+ | | 8.2k | gnd | 200k \\| | npn |-----+---+ |--' 100k \\_|_ | 1W Zener/_\\ 100V? | | | 0v-------------+-----------+----------0v
Reply to
Winfield

Methinks the 1meg is in the wrong place; how does the circuit turn on?

Reply to
Robert Baer

...try the IRFBG20, at a rather hefty 1000V rating... And they have a number of others from 600V up; look in DigiKey...

Reply to
Robert Baer

formatting link
?Ref=29770&Row=612787&Site=US

Gack! Vishay seems to be scarfing up everything! Maybe ST Micro is the way to go...

Reply to
Robert Baer

Triodes are on till you turn them off.

Reply to
John Popelish

Can you elaborate? I've never understood how foldback current limiting really works. Descriptions of example circuits in books and the 'net occasionally point out that something has foldback, but they never really explain how a circuit can have two different stable operating points at a given current, and how it makes sure to be using the right one.

Thanks, robert

Reply to
Robert Latest

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.