BA7OMW
Member
Offline
Posts: 14
|
|
« on: April 20, 2024, 11:23:59 AM » |
|
[I'm from a non-native English speaking country and my content is provided by Google Translate]
There isn't enough room where I live to install a full-size 160M band antenna, so I need a more powerful transmitter to make up for the inefficient antenna.
I originally decided to use a Class E amplifier, but a high voltage vacuum variable capacitor suitable for a high power Class E amplifier output tuner was beyond my budget.
Then I discovered W1VD’s Current-Mode Class D power amplifier design on the Internet, which uses a transformer for impedance transformation without the need for tuning components. Seems easier to build too.
After continuing to search for CMCD power amplifier related content, I found the CMCD calculator written by LU5HAH and calculated a design that seemed feasible.
My electronic knowledge is not very rich and I cannot complete the design by myself, so I hope the experts in the forum can help me. Thank you very much.
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #1 on: April 20, 2024, 11:29:52 AM » |
|
Regarding the outdated MOSFETs used in the design, NOS's FQA11N90 MOSFETs are still readily available where I live for $4.30 each. If there is a better MOSFET option please let me know.
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #2 on: April 20, 2024, 11:37:13 AM » |
|
It is found from the calculator that the output impedance at 80V carrier voltage and 625W carrier power is exactly 50Ω, which can be output through a simple 1:1 choke. From the Class E website, I learned that the safe operating current of the FQA11N90 should be around 1A, and the total current of 7.9A calculated on the calculator looks fine. But is 80V carrier voltage too high for FQA11N90?
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #3 on: April 22, 2024, 10:11:22 PM » |
|
Hi... Not sure of your name or country not that it matters.... I am not an expert on CMCD but will give it a go to help... Tonight will check though whether all your calculations of the tuned circuit are correct using 4 X 11N90s either side etc: and advise soon.... The current balun should be fine and also the DC blocking capacitors locations can be positioned before or after the balun... Using CMCD with 80 volts applied should be OK.... Driving current will be interesting but your operating frequency is 160M so thats in your favour.... Are you using Digital drive or Sinewave drive ? Wayne
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #4 on: April 23, 2024, 02:33:21 AM » |
|
Hi... Not sure of your name or country not that it matters.... I am not an expert on CMCD but will give it a go to help... Tonight will check though whether all your calculations of the tuned circuit are correct using 4 X 11N90s either side etc: and advise soon.... The current balun should be fine and also the DC blocking capacitors locations can be positioned before or after the choke... Using CMCD with 80 volts applied should be OK.... Driving current will be interesting but your operating frequency is 160M so thats in your favour.... Are you using Digital drive or Sinewave drive ? Wayne Thank you for your help. I plan to use digital drivers, using an IXDD614 to drive two FQA11N90s.
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #5 on: April 23, 2024, 05:41:25 AM » |
|
Noticed your online... The calaculator seems great... Do you have the link...
Ohhh you just departed...
Wayne
|
|
|
Logged
|
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #7 on: April 23, 2024, 06:14:02 PM » |
|
Sorry for the delay... I decided not to use LA5HAH calculator but my own method and came up with these figures... 5200pF in parallel with 1.4uH equals a resonant frequency of 1.865Mhz... Each 11N90 has 300pF capacitance which totals 1200pF... 5200-1200=4000pF for the external capacitor.... The Q is very low so the bandwidth will be quite large. The inductor will also be large but manageable and will get hot so hot that you may have to bolt it down... Have you decided on the output balun construction whether or not to use ferrite beads or a toroidal core... You could use twisted wire... Wayne
|
|
|
Logged
|
|
|
|
w9jsw
Two shots of Whisky
Contributing
Member
Offline
Posts: 599
|
|
« Reply #8 on: April 23, 2024, 06:37:16 PM » |
|
Some practical observations. 50ohms is good to make the output circuit simple, but how will you modulate the amplifier? PWM does not like to see 50 ohms. The output filter inductors will be huge. You could modulate it easier using a conventional mod transformer approach if that is your intended source impedance. Some ideas here. http://amfone.net/Amforum/index.php?topic=30798.0I have a 4 device H-Bridge making 375W at 96V. Have the rest of the circuit designed using an Antek transformer as the mod trans, but life has gotten in the way, so I have not implemented the circuit. Should work fine... 625W should be do-able. Making carrier is the easy part. Making a transmitter is more effort. Also, 625W at 80V is 7.8 amps. Peak current could be close to 31 amps. Going to take a solid power supply. John
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #9 on: April 24, 2024, 02:46:34 AM » |
|
Some practical observations. 50ohms is good to make the output circuit simple, but how will you modulate the amplifier? PWM does not like to see 50 ohms. The output filter inductors will be huge. You could modulate it easier using a conventional mod transformer approach if that is your intended source impedance. Some ideas here. http://amfone.net/Amforum/index.php?topic=30798.0I have a 4 device H-Bridge making 375W at 96V. Have the rest of the circuit designed using an Antek transformer as the mod trans, but life has gotten in the way, so I have not implemented the circuit. Should work fine... 625W should be do-able. Making carrier is the easy part. Making a transmitter is more effort. Also, 625W at 80V is 7.8 amps. Peak current could be close to 31 amps. Going to take a solid power supply. John Oh, it seems I forgot the critical PWM modulator load impedance issue. I've posted the new modified design below with a drain impedance of 12.6Ω. I plan to use the same five IRFP260N PWM modulators as the 24 MOSFET Class E transmitter. If five IRFP260N PWM modulators can drive a 24 MOSFET Class E transmitter with a 1000W carrier, then it can theoretically drive a 625W CMCD transmitter.
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #10 on: April 24, 2024, 02:51:49 AM » |
|
Sorry for the delay... I decided not to use LA5HAH calculator but my own method and came up with these figures... 5200pF in parallel with 1.4uH equals a resonant frequency of 1.865Mhz... Each 11N90 has 300pF capacitance which totals 1200pF... 5200-1200=4000pF for the external capacitor.... The Q is very low so the bandwidth will be quite large. The inductor will also be large but manageable and will get hot so hot that you may have to bolt it down... Have you decided on the output balun construction whether or not to use ferrite beads or a toroidal core... You could use twisted wire... Wayne thank you very much for your help. I just realized that the MOSFET Coss entry in the LU5HAH calculator only calculates the Coss of one MOSFETs. If you use more MOSFETs in the actual design, you need to subtract more Coss yourself. I plan to make an impedance transformation transformer and choke using 28B1020-100 cores.
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #11 on: April 24, 2024, 02:56:25 AM » |
|
I revised the design after discovering problems with the initial design. Adjust the drain impedance to 12.6Ω to suit the PWM modulator. Also increase the number of MOSFETs to 12 to amortize the increased current.
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #12 on: April 24, 2024, 04:43:33 AM » |
|
Ok thats great ..... Seems like your organised now and can proceed with the build.... Your PWM will work fine into those loads....
Let us know how things work out....
Wayne
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #13 on: April 29, 2024, 01:56:11 AM » |
|
Have you built the transmitter ?
I may have given you the wrong information regarding the value of the external parallel capacitor.... Have new figures now using a higher LC ratio.... A smaller inductor of .9uH .... That is 10T on a .750mm former and this would be easier to pull squeeze to change the inductance for resonance or maximum efficiency.... Also I used the Drain Source capacitance of 300pF which is incorrect.... It should be the Reverse Transfer Capacitance which is 40pF per FET totalling appox 160pF.... So the external capacitor should be 8000pF appox for a resonance at 1876Khz....
Wayne
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #14 on: April 29, 2024, 08:33:09 AM » |
|
Have you built the transmitter ?
I may have given you the wrong information regarding the value of the external parallel capacitor.... Have new figures now using a higher LC ratio.... A smaller inductor of .9uH .... That is 10T on a .750mm former and this would be easier to pull squeeze to change the inductance for resonance or maximum efficiency.... Also I used the Drain Source capacitance of 300pF which is incorrect.... It should be the Reverse Transfer Capacitance which is 40pF per FET totalling appox 160pF.... So the external capacitor should be 8000pF appox for a resontance at 1876Khz....
Wayne
Thanks for the new data. I am still preparing the raw materials needed to build the transmitter, and I plan to start building it after I have all the materials.
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #15 on: May 01, 2024, 10:22:32 PM » |
|
Have you built the transmitter ?
I may have given you the wrong information regarding the value of the external parallel capacitor.... Have new figures now using a higher LC ratio.... A smaller inductor of .9uH .... That is 10T on a .750mm former and this would be easier to pull squeeze to change the inductance for resonance or maximum efficiency.... Also I used the Drain Source capacitance of 300pF which is incorrect.... It should be the Reverse Transfer Capacitance which is 40pF per FET totalling appox 160pF.... So the external capacitor should be 8000pF appox for a resonance at 1876Khz....
Wayne
I have a little question to ask. If you want to add a TVS to protect the drain of the MOSFET from being damaged by voltage spikes, how many volts higher than the power supply voltage is needed for the TVS?
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #16 on: May 02, 2024, 07:58:53 PM » |
|
You would probably work on 4 times your Power Supply Voltage... You have 80volts on the drains of the 11N90s so your Supply would be 200volts .... thats around 800 volts I suppose.... In my opinion you are pushing this TX rather hard and there would not be too much room for safety.... When I test a transmitter its at a much higher level than what I would operate at.... In your case I would be hesitant to go higher etc:
Maybe others might like to comment...
Wayne
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #17 on: May 02, 2024, 10:01:57 PM » |
|
You would probably work on 4 times your Power Supply Voltage... You have 80volts on the drains of the 11N90s so your Supply would be 200volts .... thats around 800 volts I suppose.... In my opinion you are pushing this TX rather hard and there would not be too much room for safety.... When I test a transmitter its at a much higher level than what I would operate at.... In your case I would be hesitant to go higher etc:
Maybe others might like to comment...
Wayne
Insufficient safety redundancy does it mean current or voltage? Should I increase the number of MOSs or replace them with higher voltage MOSs?
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #18 on: May 02, 2024, 11:05:58 PM » |
|
Definitely keep going with your build.... Good testing will show things and you can assess as you go along.... Maybe at the moment leave the diodes out and when completed add them etc: Wayne
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #19 on: May 02, 2024, 11:32:25 PM » |
|
What FETs are you using in your PWM...
Are they IRFP260s ?
Also do you have overload protection ?
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #20 on: May 03, 2024, 01:39:00 AM » |
|
What FETs are you using in your PWM...
Are they IRFP260s ?
Also do you have overload protection ?
Yes, I will use IRFP260 in PWM circuit. The analog overload protection circuit built with op amps is a bit difficult for me to understand, I plan to use an MCU controlled overload protection circuit where the overload protection thresholds can be configured through simple programming.
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #21 on: May 03, 2024, 04:08:36 AM » |
|
Thats good regarding your overload protection.... If your planning on applying 80volts to the TX with a Mark to Space ratio of 60:40 the Power Supply will have to be 200 volts... The IRFP260s maximum is also that value....
Wayne
|
|
|
Logged
|
|
|
|
BA7OMW
Member
Offline
Posts: 14
|
|
« Reply #22 on: May 03, 2024, 08:17:46 PM » |
|
Thats good regarding your overload protection.... If your planning on applying 80volts to the TX with a Mark to Space ratio of 60:40 the Power Supply will have to be 200 volts... The IRFP260s maximum is also that value....
Wayne
Maybe replace the IRFP260 with twice the amount of IRFP460?
|
|
|
Logged
|
|
|
|
vk3alk
Member
Offline
Posts: 264
|
|
« Reply #23 on: May 04, 2024, 06:01:52 AM » |
|
Yes you could use the IRFP460..... If you used 5 in parallel like you stated a few posts before that would be fine...
Maybe for now complete the TX and worry about your PWM then...
Wayne
|
|
|
Logged
|
|
|
|
KD6VXI
Contributing
Member
Offline
Posts: 2692
Making AM GREAT Again!
|
|
« Reply #24 on: May 04, 2024, 02:17:41 PM » |
|
I don't remember the spec, but I used IRF640 in my PWM.
Cheap cheap cheap!
I also used them at 160and 80 meters.
--Shane WP2ASS / ex KD6VXI
|
|
|
Logged
|
|
|
|
|