The AM Forum
March 28, 2024, 02:55:48 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: TX / RX classe D PWM control  (Read 9363 times)
0 Members and 1 Guest are viewing this topic.
CT4RK
Member

Offline Offline

Posts: 7



« on: January 17, 2020, 07:05:25 PM »

Hi all AMs

I built a classe D 80m transmitter based in the W1VD 80m board, with 4 STW15N80K5, running 90 V drain on the modulator input, and I`m able to take more than 400W carrier, although, for security, I set it to no more than 300 / 350W carrier. Sometimes I havel faults, with MOSFETs burned in PA. The waveform on the PA gates is correct with no crossing. The modulator works with two IRFP260N, but the modulator has already burned the MOSFETs with  no apparent reason and applied the 90V directly to the PA, burning the MOSFETs of the PA. I've been watching on the internet and almost everyone cuts HV during RX, I don't have in my sequencer way to cut the HV (but I can do it trought a vacum relay)! I cut the 12V from the PWM generator to extinguish the pulses in the gates of the modulator  MOSFETs. So, the 90V HV are permanently in the drain of the modulator MOSFETs.
I would like to know if anyone can help me, and tell me what is the importance of cutting the HV in the drains of the modulator's MOSFETs, instead of cutting the gate pulses to turn OFF the MOSFETs.

TNX in advance
73 from CT4RK
 
Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
KQ6F
Member

Offline Offline

Posts: 124


« Reply #1 on: January 17, 2020, 08:21:47 PM »

There is nothing wrong with keeping the 90 volts on all the time provided sequencing is done properly.  Perhaps if you describe how the PA gate drive, modulator gate drive, and T/R relay are sequenced - both during Tx turn-on and Tx turn-off, we could possibly see the problem. 

Are the modulator MOSFETs connected in parallel or in a half-bridge configuration?

Can you be certain the failure mechanism starts with the modulator MOSFETs?  A shorted PA MOSFET could also be the first event, leading to failure of the modulator MOSFETs.

You might consider posting the schematic of your modulator.  That might help also.
Logged
CT4RK
Member

Offline Offline

Posts: 7



« Reply #2 on: January 17, 2020, 08:59:03 PM »

Hi KQ6F
TNX for your repply. The modulator MOSFET are in parallele. The sequencer are:

1º -  Antena change relay, (My rig is a transveiver not only a transmitter)
2º -  starting modulator pulses
3º -  starting the RF generator /180º shift
In the second step I have a soft start at the PWM. It starting around 5% duty cycle and go up to 45% in about 0,5 seconds.
I have also a Hall effect  current detector, adjusted for 20 Amp, and an ARDUINO based SWR detector, that shutdown the sequencer if the SWR is more than 2:1 and the current exceeds 20 Amp Any way, with this protections I have some faults.
I atach some photos of the transceiver and also de circuyt diagram of the modulator.


* gate drive.jpeg (172.48 KB, 1920x933 - viewed 733 times.)

* PA.jpeg (199.14 KB, 1920x933 - viewed 922 times.)
Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
CT4RK
Member

Offline Offline

Posts: 7



« Reply #3 on: January 17, 2020, 09:10:05 PM »

Here is the modulator diagram, and a photo of the transceiver

* PWM modulator.pdf (889.85 KB - downloaded 376 times.)

* IMG_20180624_162325.jpg (2086.59 KB, 4160x2336 - viewed 875 times.)

* PA layout.jpg (2284.66 KB, 4160x2336 - viewed 908 times.)
Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #4 on: January 17, 2020, 09:53:33 PM »

You might want to compare your modulator with this one.

http://amfone.net/Amforum/index.php?topic=42504.0

This one has been built quite a few times.

Pat
N4LTA
Logged
CT4RK
Member

Offline Offline

Posts: 7



« Reply #5 on: January 18, 2020, 09:21:39 AM »

Hi Pat
TNX for your tópics. I think that some fault with MOSFETs burnout in switched amplifiers are more or less normal, and is not a hard problem. Also the modulator worked fine during months without any faut. All  the folks are begining when I start the transmitter with wrong antenna. It is normal to burn the PA, because the 20m antenna have a very low Z on 80, and over charge the PA ( in my PA I have a 1:4 coaxial transformer rather than a convencional RF transformer. You can see the PA diagram here: ( http://www.w1vd.com/75M375WclassDRev2.0.pdf  ). I have some mods from this circuit. First I have only 4 MOSFETs with low Cin ( 1nF Cin) , second I Have a 0,5R 2W resistor in serie with gate, and finaly I have no capacitor in paralel with the shunt coil. I have the OUT capacintance from the MOSFETs and in each side two ATC100B 220pF capacitors to GND. I don`t understanding  why the IRFP260N are burned (only one) because two in paralel are able to handle more than 300 Amp peak! I`m trying understand if the 90V present in the modulator drains without gate pulses are a problem! You tell me that is not a problem and it is good news because I can save a vacum relay. Of ocurse, if the modulator go to short circuit it put 90V directly to the PA and burn the PA! Maybe the advantage to cut the 90V is if the modulator burn, and the over current act, and shutdown the 90V, the PA can survive!

73 Carlos
CT4RK
Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
KQ6F
Member

Offline Offline

Posts: 124


« Reply #6 on: January 18, 2020, 12:00:49 PM »

Carlos -

I sent you an email.

73, Rod KQ6F
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #7 on: January 18, 2020, 08:03:12 PM »

Wow !

No MOSFETs should ever burn out if everything is working and built correctly.  The modulation should have a FAST overcurrent shutdown.  The RF amplifier should have TransZorbs across the drains to stop voltage spikes.

The design should have at LEAST a 100% headroom for voltage (preferably 150% headroom), and for current the headroom should be very high as well.

With this, your transmitter should never fail.

I have class E rigs that are more than 10 years old and have NEVER had a failure.  I have transmitted into short circuit, open circuits, and everything between.  Do consider a redesign, first of the modulator, so there is an overload shutdown that works.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
CT4RK
Member

Offline Offline

Posts: 7



« Reply #8 on: January 19, 2020, 08:13:56 AM »

Hi Steve
TNX for your reply

No MOSFETs should ever burn out if everything is working and built correctly.  The modulation should have a FAST overcurrent shutdown.  The RF amplifier should have TransZorbs across the drains to stop voltage spikes.

Of course! If everything is OK nothing should be burn! Yes! my transmitter have a fast shutdown overcurrent, based in hall effect detector, between modulator and PA. The PA have tranzorbes in all drains and gates (500v in drain and 18V in gate) ! Also the four PA MOSFETs STW15N80K5 have an internal gate tranzorbe and the conventional DS tranzorbe.

The design should have at LEAST a 100% headroom for voltage (preferably 150% headroom), and for current the headroom should be very high as well.

I`m running 90V max, and the PA MOSFETs are 800V. The IRFP260N on the modulator are for 200V! I think that in classe D this ratings are not a problem. Two IRFP260N in paralel are able to handle near off 300Amp uS peak.

I have class E rigs that are more than 10 years old and have NEVER had a failure.  I have transmitted into short circuit, open circuits, and everything between.  Do consider a redesign, first of the modulator, so there is an overload shutdown that works.

OK! You are a very experienced in class E and D transmitters and have a lot of skils is this field! This is my first "switching" transmitter! I am a beginner on this kind of transmitter and it is, for me, especially useful to learn! So I looked for help with the problems on the forum! When a problem happens I don't just solve it! I always try to know why it happened! I like to learn! I'm not a radio amateur just only to speak into the microphone!I`m more Homebrewer!
Rod give me a good ideia! Maybe my problem is the TX/RX sequence! A connect the modulator in the second step and apply drive to PA in the last step! during 100mS step time, the modulator is unloaded, both in the TX/RX and RX/TX. The suddenly loaded and unloaded  the PWM filter can cause this desasters! I need change the sequence of the control! I do it today!

Best 73 CT4RK



Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #9 on: January 19, 2020, 08:25:17 AM »

Yes, sequencing is VERY important.

I just don't see how you are ever able to damage anything in the modulator  The shutdown should turn off the modulator way before anything could be damaged.

You may short a TransZorb in the RF amplifier if there is a switching transient, but I would not expect any MOSFETs to be damaged.

Unless something is hitting a gate somewhere with high voltage...

Let us know how you make out with the sequencing changes!!!

Regards,  Steve
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
vk3alk
Member

Offline Offline

Posts: 238


« Reply #10 on: January 19, 2020, 09:39:10 PM »

Its possible the Drain Source junction is failing....
Not sure if he has protection in the TX Pulse Train.....

Could be a poor design of his Class D modulator or perhaps the LPF first inductor etc: .....
The back EMF of that inductor could be causing problems.....

I have never had a IRFP260N go that way but 200 volt head room is not much if thats the case....

Only my 20 cents worth though....


Wayne


Logged
CT4RK
Member

Offline Offline

Posts: 7



« Reply #11 on: January 20, 2020, 09:59:32 AM »

Hi all
TNX for your help!
Now, after your help, especially to Rod that change with me some e-mail (tanks Rod for your atention to me), about the sequencer, and after do some mods, mainly change the operations of the sequencer, for, never leave the modulator unloaded, and add a HV vacum relay to cut the 90V during RX, I have made some tests, icluding some hard tests like overmodulation, and fast switching between TX and RX,  TX without antenna a half power etc, and everything remains OK! No trips, and no "Burn-pop-fire". Not bad!...let's see how long !!!

Best 73 CT4RK
Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
CT4RK
Member

Offline Offline

Posts: 7



« Reply #12 on: January 20, 2020, 10:08:37 AM »

Of course when I transmitte without antenna I have a "HI SWR" trip, as soon as it reaches 10W! ...Is signal  that working!
Logged

The true sign of intelligence is not knowledge, but imagination.
Albert Einstein
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 0.058 seconds with 19 queries.