The AM Forum
March 28, 2024, 05:57:32 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 ... 5 [6] 7   Go Down
  Print  
Author Topic: Board for WA1GFZ MOSFET Audio Driver  (Read 72941 times)
0 Members and 1 Guest are viewing this topic.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #125 on: May 01, 2019, 11:16:18 AM »

thanks for the tips Bob.

But why do we have to drill into the fins ?
The thickness of the sink plate, from flat to
the valley floor between fins, is 7/32”...
shouldn’t that be the drilling depth ?

Peter



Hi Peter -

If the plate is thick enuff as you say, then no problem.   You will never hit a fin.

Just to be clear for other builders: In the case of thin plates, the problem is that the FET mounting holes in the PC board were not designed for any particular heatsink fin spacing, thus some holes will probably line up over fins.  If a standard heatsink was used for all projects, like for a commercial production run, then the designer would make sure no fins needed drilling...  Wink

If you have more than one heatsink, maybe one will match up with no hassles. Pot luck.  Though, tapping into fins is normally no problem other than a stuck drill.

BTW, right now there is a lot of email traffic going on about testing the board for performance and optimizing parts selection. These results will be posted in the future.


T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
KC2ZFA
Member

Offline Offline

Posts: 441



« Reply #126 on: May 01, 2019, 11:56:29 AM »

thanks Tom.

to all, once completed can the board be tested using resistor loads to represent the tubes ?
If yes, how does one determine the resistor values for a given tube ?

And by tested I'm assuming it means putting in audio sigs from a generator and scoping various places to look for correct pk2pk, distortion, parasitics, etc...

Peter
Logged
W1RKW
Contributing
Member
*
Offline Offline

Posts: 4410



« Reply #127 on: May 01, 2019, 04:48:20 PM »

Hi Peter,
Sorry for the late response.  I was saying you need to avoid grazing the fin as the drill bit penetrates the flat which is what I did with one hole and broke the bit because it jammed.  If you can get all the holes centered in a manner where a drill bit does not graze a fin that would be ideal.  If you drill straight into a fin that is OK too as Tom mentioned.  

Received the transformer and the 680uF caps today.
Logged

Bob
W1RKW
Home of GORT. A buddy of mine named the 813 rig GORT.
His fear was when I turned it on for the first time life on earth would come to a stand still.
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #128 on: May 05, 2019, 08:08:14 AM »

BOM and schematic error - C11 needs to be .01uf.

* BOM_WA1GFZ Mosfet Audio Driver.xls (10 KB - downloaded 350 times.)
* New WA1GFZ Mosfet Audio Driver.pdf (61.55 KB - downloaded 345 times.)
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #129 on: May 05, 2019, 11:02:58 AM »

An option for those who can't abide a fin in the way, or where it's half on and off the hole and difficult to drill for tapping, would be a Dremel tool. a 'worn down' cutting disc will slice off the fin base usually.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #130 on: May 10, 2019, 07:01:51 AM »

Hot off the press - Frank's comments after the build and testing in his Valiant.

News is that the board has no errors and the components are all correct (after the C11 change to .01uf).

good stuff!

For demanding applications, a regulated supply is suggested. I am toying with the idea of making some regulator module PCBs based off a design that Phil posted on a thread a while ago. The concept would be to put all components on a board and use 2 of them for the two rails. The populated boards plus the transformer is all that would be needed. No other discrete parts.

The schematic of the full supply is probably not correct. I think the 230V trans will only do around 280V regulated. Probably need a 250V trans for a solid 300V regulated.

I did not change any values from Phil's schematic. I am sure some values will need change. This is all preliminary work on the concept of an easy regulated supply. Making PCBs is easy.

John

* HV Regulator with Zeners.pdf (54.03 KB - downloaded 351 times.)
* Schematic_WA1GFZ-Mosfet-Audio-Driver_Regulated-Supply.pdf (52.59 KB - downloaded 360 times.)
* Schematic_Regulator_Component.pdf (43.31 KB - downloaded 326 times.)
Logged
W1RKW
Contributing
Member
*
Offline Offline

Posts: 4410



« Reply #131 on: May 10, 2019, 04:51:57 PM »

I was planning on doing a basic shunt regulator with zeners alone if some sort of regulator is needed. But I won't dive into that until the fall season.  I'm on hold for any radio projects other than the crystal set I'm still trying to get through.  

John, thanks for the new files. All is archived for future use.

Tom, I'll have some results in response your last question some time this weekend. Weather is supposed to suck so I'll be at the bench and operating position.
Logged

Bob
W1RKW
Home of GORT. A buddy of mine named the 813 rig GORT.
His fear was when I turned it on for the first time life on earth would come to a stand still.
KK4YY
Member

Offline Offline

Posts: 565


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


« Reply #132 on: May 10, 2019, 06:25:43 PM »

I think you need to revisit the way the bridge rectifiers are drawn on
"Schematic_WA1GFZ-Mosfet-Audio-Driver_Regulated-Supply.pdf"

Just sayin'. Wink


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
K8DI
Contributing
Member
*
Offline Offline

Posts: 423


« Reply #133 on: May 10, 2019, 09:33:52 PM »

For your schematics, you've kept the resistors in front of the zener the same as the source schematic -- but it was dropping a hundred volts across the pair of resistors. That means several milliamps, most likely above the Izk specification of whatever diodes he was using. Izk is the minimum amount of current you need through a zener to have it regulate.  With dropping the ten volts shown in your schematic, the current is just under half a milliamp.  Typical diodes (On semi, part numbers chosen by sticking 'though hole', '100v', '5w', and 'in stock' on Digikey's website) have Izk values of 1mA. You'll want to lower the resistor values to get a couple mA through the diode stack, or raise the input voltage to at least 30-40 v above the zener voltage.

Ed

* 1N5333B-D.PDF (130.02 KB - downloaded 312 times.)
Logged

Ed, K8DI, warming the air with RF, and working on lighting the shack with thoriated tungsten and mercury vapor...
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #134 on: May 11, 2019, 02:23:23 AM »

For your schematics, you've kept the resistors in front of the zener the same as the source schematic -- but it was dropping a hundred volts across the pair of resistors. That means several milliamps, most likely above the Izk specification of whatever diodes he was using. Izk is the minimum amount of current you need through a zener to have it regulate.  With dropping the ten volts shown in your schematic, the current is just under half a milliamp.  Typical diodes (On semi, part numbers chosen by sticking 'though hole', '100v', '5w', and 'in stock' on Digikey's website) have Izk values of 1mA. You'll want to lower the resistor values to get a couple mA through the diode stack, or raise the input voltage to at least 30-40 v above the zener voltage.

Ed

If you are referring to the schematic below let's look at the numbers.

3X200V zeners  = 600 volts + 1X100V zener = 700 volts. Source voltage is 800 volts.

800-700 volts = 100 volts. 100 volts/20.2k = 5 mA.

So 5 mA is the current going through the series string of 2.2k, 18k, and the 4 zener diodes.

Look at this reference Page 6;

https://www.onsemi.com/pub/Collateral/1N5333B-D.PDF

The curves start at 0.1 mA and goes to some max current. This means the zener starts regulating (without excessive switching noise) at a minimum current of 0.1 mA. So Izmin = 0.1 mA.

Where is the curve the most linear? About 5 mA. What is the current flowing through the series combination of the 2.2k, 18k, and the 4 zener diodes?

5 mA.

The power dissipation of each 200 volt zener is 5e-3X200 V = 1 Watt. What is the maximum rated power of a 5 Watt zener? 5 watts. So we have about an 80% margin of safety in terms of power dissipation for the 200 volt zener.

The power dissipation of the 100 volt zener is 5e-3X100 V = 0.5 Watts. What is the maximum rated power of a 5 Watt zener? 5 watts. So we have about a 90% margin of safety in terms of power dissipation for the 100 volt zener.

Conclusion: One has to examine and study the device curves, whether it be solid-state or Thermionic-State, to see the whole picture.

Izt is NOT the minimum current at which a zener can regulate. Izt is the "Test" current for the zener specifications and Izt varies with zener voltage and power rating.


Phil - AC0OB

 

 



 


* HV Regulator.pdf (47.43 KB - downloaded 333 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 #135 on: May 11, 2019, 07:51:41 AM »

Thanks Phil, I always appreciate your in-depth explanations of how a circuit works.

Thanks for all the comments on the regulator. Going to drop the zener to 280V. That gives me 1.5ma thru the zener. Looking at the 150 and 130V curves that is within the linear portion, I believe. 1K/9K would bump it up to 3ma.

And fixed the booboo on the bridges... Ignore the 1n4148 diode in the schematic - that was the default footprint. Will use 1N5408 or similar...

Forgot to add the new design notes.

John

* Design Notes - 20190510-1.pdf (2549.64 KB - downloaded 363 times.)
* Schematic_WA1GFZ-Mosfet-Audio-Driver_Regulated-Supply.pdf (90.99 KB - downloaded 342 times.)
Logged
KK4YY
Member

Offline Offline

Posts: 565


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


« Reply #136 on: May 11, 2019, 09:33:44 AM »

Phil,

I looked at those same graphs a few weeks ago when designing a dual voltage regulated supply. The log scale was hard to read, and I had to zoom in tight, to see it clearly. I found that figures 7,8, and 9 use three different log scales to cover the range of nominal zener voltages! Very deceiving.

I'm not trying to be nit-picky, really. But the way the data is displayed on that sheet is terrible.


Don


* bad graphs.png (16.51 KB, 444x421 - viewed 623 times.)
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #137 on: May 11, 2019, 11:17:37 AM »

prelim pcb. Need to shrink it more.

* PCB_Reg0PCB_20190511101116.pdf (5.79 KB - downloaded 343 times.)
Logged
KC2ZFA
Member

Offline Offline

Posts: 441



« Reply #138 on: May 11, 2019, 11:42:46 AM »

constant load of 100 mA per 300vdc rail, why can't I regulate this with gas regulators ? 4 0D3's in each rail (two in parallel in series with two more in parallel). I think each 0D3 can handle 50 mA. Talking outa my behind here ?

Peter
Logged
KK4YY
Member

Offline Offline

Posts: 565


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


« Reply #139 on: May 11, 2019, 04:31:56 PM »

John,

I bet your dual rail regulated power supply is going to look better than mine does! Roll Eyes


Don


* dual rail ps_sm.png (831.13 KB, 614x461 - viewed 679 times.)
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #140 on: May 11, 2019, 06:50:32 PM »

This is what I will order. Will be able to fit a pair on a 100mmx200mm heatsink.

I use EasyEDA for the layout. They have an initial sales offer to build 10 of them for $5.00 plus free shipping. Going to give them a try.


* PCB_Reg0PCB.pdf (16.93 KB - downloaded 324 times.)
* Schematic_Regulator-Component_V_1.pdf (45.03 KB - downloaded 337 times.)
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #141 on: May 11, 2019, 06:59:18 PM »

Thanks Phil, I always appreciate your in-depth explanations of how a circuit works.

Thanks for all the comments on the regulator. Going to drop the zener to 280V. That gives me 1.5ma thru the zener. Looking at the 150 and 130V curves that is within the linear portion, I believe. 1K/9K would bump it up to 3ma.

And fixed the booboo on the bridges... Ignore the 1n4148 diode in the schematic - that was the default footprint. Will use 1N5408 or similar...

Forgot to add the new design notes.

John

Hi John, getting closer and lookin' good.

Below is my suggestion for the regulators with 310V in and 280V out.

Your 68k bleeder resistor would dissipate 1.2W so my suggestion would be to raise it to 220k 1W for a bleeder current of 1.3 mA and a power dissipation of 400 mW.


Phil - AC0OB  


* 280V Regulator.pdf (51.43 KB - downloaded 326 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 #142 on: May 11, 2019, 07:04:55 PM »

Thanks Phil!

I see that I forgot Cf.

Also, do you think it much better to include the fuse on the input to the FET after the diodes? I was planning on only fusing the primary.

Thanks for the revisions. You are welcome to have a few free boards when they arrive.

What does your model show for ripple?

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

Posts: 1768


« Reply #143 on: May 11, 2019, 07:30:45 PM »

Thanks Phil!

I see that I forgot Cf.

Also, do you think it much better to include the fuse on the input to the FET after the diodes? I was planning on only fusing the primary.

Thanks for the revisions.

John

There advantages and disadvantages to each, but it would seem to require two fuses either way.

I think the main advantage of fusing each 310V PS section is in troubleshooting.


Phil
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 #144 on: May 11, 2019, 08:23:43 PM »

deleted - not adding a fuse per rail. If a fuse blows we don't want the other rail to still be driving the modulator. Best that both go away, per Frank.
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #145 on: May 12, 2019, 02:24:48 PM »

deleted - not adding a fuse per rail. If a fuse blows we don't want the other rail to still be driving the modulator. Best that both go away, per Frank.

For the SS driver putting one fuse in the T3  primary IS probably the best way to implement protection.  Cool.

https://www.youtube.com/watch?v=pjW_z02EjaA   Grin


Phil
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 #146 on: May 12, 2019, 10:11:53 PM »

Phil, do you have any simulation data? What will the ripple look like?
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #147 on: May 13, 2019, 12:21:52 AM »

constant load of 100 mA per 300vdc rail, why can't I regulate this with gas regulators ? 4 0D3's in each rail (two in parallel in series with two more in parallel). I think each 0D3 can handle 50 mA. Talking outa my behind here ?

Peter

Gas regulators - 0D3 iIRC it is the Octal version - - can do 40mA maximum, and for good life is better to run lower. The obstacle you will face is that the firing voltage is higher than the regulating voltage, so in simple parallel they won't all fire. One will, then the regulation voltage across them all will be too low for the other(s) to fire.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #148 on: May 13, 2019, 02:37:48 AM »

Phil, do you have any simulation data? What will the ripple look like?

Hi John,

On the regulator with the turn-on time constant R1, CTC? I did not run one but I did run one and implement the same type of circuitry shown below (for your implementation Sht. 1 ) for an EF Johnson modulator screen regulator in which the input voltage was 800 volts and the output voltage was 300 volts; the output ripple was less than 50 mV under load.

How much ripple can you tolerate? How much total gain do you have and how much will that gain amplify the ripple at the grids? Can the +, - power supply transformer secondaries be phased such that much of the ripple could be cancelled at the M3, M4, and M5 stages?

If you want the lowest ripple for a simple zener regulated "pass" implementation, you will have to dispense with the turn-on time constant circuit (which I thought you still wanted) because the time constant circuit R1, CTC will not allow the zener impedance to follow the input voltage excursions. You want the zener impedance to change in real time with the excursions of the input voltage waveform and not be delayed by a time constant circuit.

If you need a lower ripple voltage, then I think a regulator with a feedback amplifier (Sht. 2) would be needed and that would increase complexity.


Phil




 

* 280V Regulator.pdf (82.37 KB - downloaded 326 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 #149 on: May 16, 2019, 07:48:43 AM »

I have extra boards on the way so I will build one of each to see. Yes, I did mention the desire for a soft start, however that may no longer be necessary.

50mv is quite good.

Transformer secondaries can be phased. They are completely independent.

Here is the final schematic/bom and the board that I ordered. I kept the 130V and 150V Zeners in the BOM as 140V Zeners were not available from Digikey.

John

* BOM_Regulator Component.xls (9 KB - downloaded 302 times.)
* Schematic_Regulator-Component_V_1.pdf (45.16 KB - downloaded 316 times.)
* PCB_Reg0PCB.pdf (17.31 KB - downloaded 331 times.)
Logged
Pages: 1 ... 5 [6] 7   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 0.066 seconds with 19 queries.