The AM Forum
March 28, 2024, 02:13:07 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Adapting a high voltage power supply regulator from a past thread  (Read 5792 times)
0 Members and 1 Guest are viewing this topic.
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« on: October 20, 2019, 07:15:58 PM »

I am attempting to modify this regulator circuit that was published in this forum last year for use as a screen regulator. I plan to use a heavy duty 60 amp 250 volt MOSFET and a 150 volt zener diode string. Input voltage will be 175 volts and an output voltage of 150 volts. Obviously the resistor values in the zener string must change. Any ideas for modification? Thanks in advance.

I would probably start off with R1 at 620 ohms and R2 and 4.4K.

Pat


* amforum regulator.PNG (151.62 KB, 910x515 - viewed 270 times.)
Logged
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #1 on: October 20, 2019, 07:42:18 PM »

Awesome!


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #2 on: October 20, 2019, 07:52:11 PM »

The power supply is complete and delivers about 176  volts with only the bleeder resistor and at about 157 volts at 3.3 amps load. This will be very similar to VE3ELQs regulator as he has described. It will operate a 450 watt carrier level  pulse width modulator. Nigel described  his regulator as very similar. I will probably adjust the output voltage to limit dissipation from no load to full load. The FET I am trying is the same one Nigel suggested - an  IRFB4332PBF  - 250 volt 60 amp  390 watts at 20C and 200 watts at 100C.  I am thinking about 3 amps at a max of 4-5 volts at full load so a decent heatsink will be needed.

Pat
N4LTA
Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #3 on: October 20, 2019, 08:00:04 PM »

Pat,

I made some PCBs for this circuit. Once you get the details ironed out on values, you are welcome to one. I added a bridge rectifier and an on-board cap to make it easy. PCB is 3.25in by 3.75in.

Here was my circuit for a 285V regulator. Phil helped me on the values.

John

* Schematic_Regulator-Component_V_1.pdf (45.16 KB - downloaded 134 times.)

* Reg-Board.JPG (155.1 KB, 568x757 - viewed 242 times.)
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #4 on: October 20, 2019, 08:17:03 PM »

Thanks John. I have already made a simpler PCB based on Nigel's description and a couple of schematics on the net. I etched a couple and burned the traces off both of them this afternoon while probing around. 175 volts and an 18000 uF filter cap makes a big POP and does a number on a PCB Board trace. The original circuit had no voltage drop at all , as if the MOSFET was turned full on.

A nice looking board - How much current  and what was the voltage across the MOSFET?  I'm Looking at 3-4 amps so a large heatsink will be needed

I just finished designing one based on the schematic above and will etch it tomorrow. I,  like you have been laying out boards for a long time and make them at home. I appreciated your offer though. Thanks for the offer.


Pat
N4LTA
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #5 on: October 20, 2019, 09:37:27 PM »

Here is a suggestion but I would use T0-247 packaged NMOS to get better heat spreading


Phil - AC0OB

* LV Regulator.pdf (43 KB - downloaded 146 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #6 on: October 20, 2019, 09:49:42 PM »

Phil

That is very close to what I tried today except for the .01uF cap. I had something like 10uF and I didn't have the protective 15 volt zener. What does Cf do. I assume it is a feedback device.

Also - what was the purpose of the series resistors ahead of the zener string rather than one resistor as you current schematic shows.

I don't expect 75 watts dissipation -because the supply squats to 156-157 volts when keyed. It is about 176-177 no load with only the bleeder. I am thinking max dissipation would be about 6 volts x 3 amps or 18 Watts which is much easier to deal with.

Thanks for your help.

Pat
N4LTA
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #7 on: October 20, 2019, 10:12:37 PM »

Phil

That is very close to what I tried today except for the .01uF cap. I had something like 10uF and I didn't have the protective 15 volt zener. What does Cf do. I assume it is a feedback device.

Also - what was the purpose of the series resistors ahead of the zener string rather than one resistor as you current schematic shows.

Thanks for your help.  Pat
N4LTA

Hi Pat,

Cf simply kills any switching noise across the zener string so it doesn't show up as noise in the output.

Since the resistor needed is not high power, only one 1/2W resistor is needed.

Since zener current is 5 mA, (175V-150V)/.005 = 5k or 4.7k.

(0.005A)^2*4.7k = 118 mW. A 1/2W resistor = 500 mW.

Hope that helps.


Phil - AC0OB


Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #8 on: October 20, 2019, 10:42:28 PM »

Thanks Phil.

I used a 2200 ohm resistor in my tests today. I was concerned that when the supply dropped to 157 volts on key down, that I might not get enough current to  get the zener diodes stable.

Would my having a large electrolytic instead of a .01uF cap have caused a problem in my test? Or not having the 15 volt diode in the circuit? I assume the 15 volt zener is a protective devicd?

I got the same voltage on the source as at the input. I thought this part of the build was going to be simple - I am probably doing something stupid! Always something!

Thanks again

Pat
Logged
MikeKE0ZUinkcmo
Member

Offline Offline

Posts: 444



« Reply #9 on: October 21, 2019, 01:07:51 AM »

You need the Zener across gate to source, and make sure its voltage is less than the max listed in the FET's data sheet.   Since this is a pulse modulator, you're going to have some transients of perhaps several volts from the load, so I'd put a bypass/filter cap from source to ground and a series diode in the output lead as well.

Logged

Mike KE0ZU

Bold Text and PICS are usually links

https://mikeharrison.smugmug.com/
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #10 on: October 21, 2019, 01:39:45 PM »

Thanks Phil.

I used a 2200 ohm resistor in my tests today. I was concerned that when the supply dropped to 157 volts on key down, that I might not get enough current to  get the zener diodes stable.

Would my having a large electrolytic instead of a .01uF cap have caused a problem in my test? Or not having the 15 volt diode in the circuit? I assume the 15 volt zener is a protective devicd?

I got the same voltage on the source as at the input. I thought this part of the build was going to be simple - I am probably doing something stupid! Always something!

Thanks again

Pat

I would keep zener current at 5 mA in order to keep zener heating to a minimum when your input voltage rises. So at 7V delta instead of 25V, you would need about (157-150)/.005 = 7V/.005 = a 1.4k resistor to keep the zeners lit.

The 15V Gate-to-source zener is to protect the gate from transients which might exceed the gates insulation or "punch through" voltage, and is a MUST.

A 10 uF cap across the zener string is way too large as you only need a cap value to kill any switching noise. In fact, too large a cap across the zener string will NOT allow the zener to follow input voltage excursions (120 Hz pulses).

It sounds as if your pass NMOS may be shorted. When handing any MOSFET, ground your wrist to avoid any HV static discharges to the device you are handling.

Also, when measuring FET terminal voltages, discharge the DMM or scope probe to ground before moving from one node to another node.

As per Mike's theory that you might have switching transients present, I have added three extra 6A 1000PIV diodes to kill any transients that might be present.


Phil - AC0OB

* LV Regulator.pdf (53 KB - downloaded 138 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #11 on: October 21, 2019, 07:29:30 PM »

Phil

Thanks for all of your help. I have etched a couple of new boards and plan to build one tonight and do some testing. I'll let you know how things work out.

Pat
N4LTA

Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #12 on: October 23, 2019, 05:44:15 PM »

Spent some time today updating my circuit board for the regulator adding a protective Zener diode and three 6 amp diodes. I changed the board to use a T0247 device to allow for better cooling.

I found that  Newark was discounting the INFINEON  IRFP4229PBF which looked like a good candidate. It is rated at 250 volts and 44 amps.

At $1.39 each - I bought ten - so if anybody tries this circuit - I'll send you one.

The other board is a VFO with digital dividers to run the Class D transmitter on 160 or 80 meters

Pat
N4LTA


* IMG_3815.jpg (527.09 KB, 2016x1512 - viewed 218 times.)
Logged
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #13 on: October 23, 2019, 09:38:40 PM »

After searching around a bit, I chose the STP7N95K3 Power MOSFET from STMicroelectronics for regulator use (and bought 10 of them). It has internal back-to-back zeners protecting the gate from ESD during handling. This also eliminates the need for an external zener protection diode in the final circuit.

I had other applications in mind too, and wanted a low input capacitance. Input capacitance seems to track the current rating of the device so I tried to minimize those parameters in favor of a higher voltage rating. The STP7N95K3 is rated at 950V and 7A with an input C of 1031pf. I doubt that I'll ever need 7A of current in any tube-type rig I may build, but 950V I just might come close to.

It may not be the perfect 'fet for everything, but at least it won't short-out when I touch it.  Cry
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #14 on: October 24, 2019, 06:16:11 PM »

Got the Voltage Regulator board assembled and received the new MOSFET. It all seems weel. I have something that I can work with now.

With two 5 watt 75 volt Zener diodes, I get 147 volts out. I have a third zener position on the board shorted with a jumper. I can add a zener there to raise the voltage to what suits the power supply droop and reduce the MOSFET power dissipation.

Thanks everyone for helping. If anyone wants to duplicate the circuit, I have some spare MOSFETS free to a good home.


* regulator.jpg (548.11 KB, 2016x1512 - viewed 249 times.)
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #15 on: October 24, 2019, 06:56:49 PM »


...With two 5 watt 75 volt Zener diodes, I get 147 volts out. I have a third zener position on the board shorted with a jumper. I can add a zener there to raise the voltage to what suits the power supply droop and reduce the MOSFET power dissipation...
 

Glad you have it working. Allowing space for a third zener to adjust final voltage was an excellent move. Cool

Your FET and D1 together must have a total voltage drop of about 3 volts. Adding a 3.3V 1 to 5W zener (1N4728 or 1N5347B) should fix it for you.  

In the future you may want to heatsink the FET.


Phil - AC0OB

* LV Regulator.pdf (47.7 KB - downloaded 129 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #16 on: October 24, 2019, 07:21:48 PM »

Pat,

If you look at my board you can see that I place a hole and mount the fet underneath. Works well to heat sink the fet on an under mount heat sink.

Any chance you can mail me a transferable reverse sheet of the vfo? I assume that is the process you use to create the pcb (laser printer transfer). Would like to try your vfo layout. I used to etch boards back when I was a kid.

No need to share the schematic, I can figure it out.

John
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #17 on: October 24, 2019, 08:31:38 PM »

John

I use the positive photo resist method mostly and sometimes I laminate FR4 with negative resist and do it that way. Negative is easier but the plots on clear film are harder to do.   - Send me your address and I'll send you a board - I'll make one the next time I do boards and send it to you. That will be easier than you trying to etch a one off board. I do this pretty often and it would just take a few minutes extra when running a board for me. The schematic is in the W1VD website.

Pat

Phil - I will definitely need to heatsink the MOSFET
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #18 on: October 26, 2019, 03:05:02 PM »

Gave the regulator a stiff test this afternoon. I attached a heatsink  and wire up a dummy load - a heavy duty 200 watt 50 ohm wirewound resistor - Just 200 watts but beefy.

At approximately 3 amps - the heatsink got warm with a 30 second key down - I dared not go more for fearing the big resistor would burn out. Voltage was reasonably stable within a volt.

It appears at 3 amps the regulator and series diode have a bit of voltage drop - about 3 volts. The delta voltage from input to output at 3 amps is about 10 volts or a power dissipation of close to 30 watts. I am going to up the zener string to get the regulated output at full load closer to 153 volts or so and lower the input to output delta voltage to reduce the dissapation of the MOSFET
to 12 watts or so.

If I can find a large 150 ohm or so power resistor I may run it for a few hours and see how it does.


* test2.jpg (605.59 KB, 2016x1512 - viewed 253 times.)
Logged
MikeKE0ZUinkcmo
Member

Offline Offline

Posts: 444



« Reply #19 on: October 26, 2019, 11:29:04 PM »

I used a similar regulator a couple years ago in my Viking Ranger for both the High and Low B+.


* +HV FnlCrop.jpg (233.6 KB, 1469x1225 - viewed 300 times.)
Logged

Mike KE0ZU

Bold Text and PICS are usually links

https://mikeharrison.smugmug.com/
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.061 seconds with 19 queries.