The AM Forum
December 02, 2024, 11:54:22 AM *
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: Nigels PWM  (Read 15194 times)
0 Members and 1 Guest are viewing this topic.
vk3alk
Member

Offline Offline

Posts: 263


« on: May 26, 2018, 10:55:22 PM »

The LTC6992-1 arrived and so completed Nigels PWM...
All went well and no problems at all....
Just have to be mindful of the audio input levels thats all.....

Nigel recommends the PWM generator circuitry be away from the output board currents but will leave that up to future builders etc:....

Have attached 3 photos...one of the quickly assembled unit....the output from the LTC6992 and the output from the LPF with an audio tone....
Also ran an audio sweep with 100 volts to the board and all went fine....

Its probably only me but are not keen on using motor controllers as PWM generators...
So if anyone is interested will build one of my own and replace the existing one on Nigels Board and post the results......


Wayne


* Nigels Board.JPG (411.97 KB, 1600x1200 - viewed 1108 times.)

* Nigel LPF Output.jpg (105.78 KB, 800x480 - viewed 984 times.)

* Nigel PWM Output.jpg (93.91 KB, 800x480 - viewed 985 times.)
Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 599



« Reply #1 on: May 27, 2018, 08:04:58 AM »

What did you use for a load on this test?

John
Logged
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #2 on: May 27, 2018, 08:22:27 AM »

Wayne,
Good work, love to see folks experimenting and building.  Scope pics look great. That LTC6992 PWM chip is very versatile and the best solution I have found to date. You can try different divide ratios by changing the V on pin 4, put a pot on there and see how it changes. And change freq with the R on pin 3. It is capable of a huge range with excellent linearity. I keep the Pulse gen board and Output board separate to keep switching transients out of the low level audio stuff, its a nasty environment in there with fast high current pulses.  I use the LTC6992 Dash 4 version which limits minimum PW to 5% and with my natural voice asymmetry going positive and slight negative peak clipping which gets filled in by the 5% minimum to keep the carrier alive, I run about 375W carrier and get 1750W PEP at full mod with no noticeable distortion. Its an easy to do mild negative peak limiter.

You may want to explore using the SI8234AB-D-IS half bridge driver feeding a pair of IRFB4227 FETs. It employs a single PWM input so no inverter is needed and has programmable dead time, I use 1K R on pin6, so it drives the FET gates directly for faster switching. See spec sheet figure  4.1B. Add .47 MFD and .01 MFD bypass caps in parallel right up tight on the FETs to prevent pulse ringing.  Best Ive tried so far.

Keep building, 73s  Nigel

https://www.silabs.com/documents/public/data-sheets/Si823x.pdf
Logged

A "Tube Free" zone.
vk3alk
Member

Offline Offline

Posts: 263


« Reply #3 on: May 27, 2018, 08:50:07 AM »

Hi John...

The output load was appox 7-8 Ohms although the LPF is designed for 12 Ohms....
The cutoff I use is fairly high at appox 30Khz....
Most of my TXs are 12 Ohms.......

Hi Nigel...

There are so many good FETs around now ... most in TO220 packages  Shocked
Can get the IRFB4227 locally here in Aus but not the driver chip....
I do buy from Mouser from time to time and they have them....quite low priced as well...

Have already made the PWM generator board so will build and post results just for interest sake....


Wayne

Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 599



« Reply #4 on: May 27, 2018, 11:00:12 AM »

I threw together a PCB for this circuit and just received them yesterday. Hope to assemble one and give it a try. However, I am guessing I will run afoul of this issue -

"I keep the Pulse gen board and Output board separate to keep switching transients out of the low level audio stuff, its a nasty environment in there with fast high current pulses." said Nigel.

Anyway it is easy enough to not populate the output part and do it dead bug if this proves to be an issue. Heck, they made me 8 boards for the total sum of $12 so I could use 2 with some interconnects and that could work.

Not working on this now however, since farming is the name of the game at this point in the year. If it ever rains I may get a chance to fire up a soldering iron.

John W9JSW
Logged
vk3alk
Member

Offline Offline

Posts: 263


« Reply #5 on: May 30, 2018, 06:21:03 PM »

Well finished the little PWM Generator Board...
Are very pleased with its performance...

Not trying to make changes to Nigel circuit as his design works fine and anyone wanting to built it should stick to his circuit...

Are posting these pictures only for general interest thats all...

Will install the PWM board on the main modulator board....
Hasn't got a TX to drive as yet so for the moment will sit on the shelf somewhere... Grin

An off set of half Vcc ( 6 volts ) is used because the opamp error amplifier cannot swing rail to rail....


Wayne


* Comparator -.jpg (102.32 KB, 800x480 - viewed 943 times.)

* Comparator +.jpg (101.64 KB, 800x480 - viewed 964 times.)

* PWM Gen Board.JPG (394.12 KB, 1600x1200 - viewed 1078 times.)
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2690


Making AM GREAT Again!


« Reply #6 on: May 30, 2018, 06:58:03 PM »

Great use of clip leads.  Class E Gods are smiling!

--Shane
KD6VXI
Logged
M0VRF
Member

Offline Offline

Posts: 177


« Reply #7 on: June 05, 2018, 05:42:29 PM »

Oh, A new thread...!

Please let me know if you get it to work.

Gone a bit quiet?

OK re the previous '350W 3 band Class D FET AM Transmitter' article related to this design, the technical issues I had were never resolved and I can see from the first images in this post the same 'Issue' remains, however I may be wrong, so please 'Enlighten' me......

In the first post (here) the p-p O/P is ONLY 62V whereas the rail is set at 100V.

ANYONE? (Nigel?)

JB.
Logged
vk3alk
Member

Offline Offline

Posts: 263


« Reply #8 on: June 10, 2018, 01:50:15 AM »

Well decided to cut Nigels PWM Generator from the main board.
Only did this to save it as a working test generator for future work...
Connected my PWM Generator and tested....all working correctly....same as Nigels....

Not quite sure what you mean there John...
In the first picture the voltage applied to the Modulator was appox 80 volts...
Using a MS ratio of 60:40 the carrier / output voltage was appox 30 volts....
Applying a synchronous audio waveform around the 30 volts means that the positive peak is appox 60 volts.....and the negative peak 0 volts....

I'm going to remove the output cores as they sing a bit and fill them with hot glue....

Also may do a modification removing all dead time .... to do this you have to create shoot through...
If you look at the output circuit you will see 2 wire links and some holes....
Will add more components to control this and mods also to the gate circuit of the FETs...

Wayne







* MyPWM.JPG (496.38 KB, 1600x1200 - viewed 1153 times.)

* MyPWM1.5Khz.jpg (106.03 KB, 800x480 - viewed 931 times.)
Logged
M0VRF
Member

Offline Offline

Posts: 177


« Reply #9 on: June 10, 2018, 03:45:58 AM »

Thanks for the 'Clarity' Wayne.

Using the LTC6992 and with a Voltage of 80 volts supplied to the modulator were you able the get 80 volts from the output when modulated to the point of clipping when run from a resting 'Carrier' ms ratio of 60:40?

Regards.

JohnB.
Logged
vk3alk
Member

Offline Offline

Posts: 263


« Reply #10 on: June 10, 2018, 04:14:34 AM »

Yes but with asynchronous waveform not synchronous .....or using a negative limiter like Steve QIX circuit to cut the negatives.......
Don't really like the way you word it as its just peak waveforms but yes...
You don't actually see 80 volts using a volt meter....


Wayne
Logged
M0VRF
Member

Offline Offline

Posts: 177


« Reply #11 on: June 10, 2018, 04:21:30 AM »

A volt meter! Huh

Using a scope measure the p-p.
Logged
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 0.043 seconds with 18 queries.