Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Dec Tue 12, 2017 7:06 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 8 posts ] 
Author Message
 Post subject: Why not a PWM xmtr project?
PostPosted: Nov Wed 22, 2017 3:53 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 664
Location: Louisiana
With all the talk of AM transmitter kits and projects, why hasn't any Pulse Wave Modulation projects been attempted or proposed?
Or did I miss them?


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Wed 22, 2017 4:58 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2073
Location: Saskatoon
I tried one several years ago. The results were less than spectacular, but I didn't put much effort into it, and might have done better with faster switching power FETs. I've been thinking about giving it another try.


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Wed 22, 2017 5:29 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 176
Location: Arlington, TX, USA
What kind of PWM are you talking about exactly? Is it just a Class D modulation amp driving a Class C final, or something different?

I've seen some designs that are called Class D transmitters, but they look to be just a push pull switching design. I always thought Class D meant a stage that uses a high frequency PWM signal and uses a low pass filter on the output to reconstruct the original low frequency signal. I'm thinking Class D at RF would use a higher switching frequency than the intended carrier frequency. But I could be wrong.

I don't see much advantage to a Class D modulation amplifier at Part 15 power levels. It's more suited to high power transmitters where the higher efficiency translates to lower power consumption and less waste heat.

Class E is not a PWM scheme; it's more of a resonant switching power supply circuit using a variable supply voltage for modulation input.


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Wed 22, 2017 9:24 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 664
Location: Louisiana
To be honest, I don't specifically understand how PWM creates an AM signal, but what I know is that the modulation determines the duty cycle of the square wave. Then on the output you have a filter network to remove the switching frequency, and you are left with AM.

I know the Harris MW-5 transmitter (used by many AM broadcast stations) used PWM to achieve high efficiency.
My RF mentor David, W5CKP, would be able to explain it to me, but he pasted away several years ago.


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Thu 23, 2017 1:16 am 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 282
Location: pensacola fl
Hi. I am an engineer who has a harris MW5 still on the air as a main tx. The PDM is an efficient modulator but it achieves it's performance of hi modulation depth and low distortion due to the fact it has negative feedback around the modulator and final as well as carrier shift control. The system is actually not that difficult to use but it has some specialized components. For part 15 the inefficiency of the modulator is not a problem. The MW5 is a class C plate modulation scheme. Jist arranged differently. The PDM filter is needed to keep the high frequency switching signal from causing interference. A properly implemented class A series modultor will work well and far less to build unless you just want to do it to do it with a PDM. The MW5 uses a 70KHZ oscillator to start with. It then feeds this into a transistor comparator. Mixed in with the audio and the sawtooth 70KHZ is a D.C. reference to set power level and to allow carrier shift correction. The output of the directional coupler is what feeds the power output meter as well as the negative feedback loop for the audio and the carrier shift correction loop. A carefully adjusted MW5 can be quite impressive sounding but they do need upkeep as tubes age and so on. The B+ suply is quite high as the modulator has to switch on and allow the capacitor at the end of the filter to charge or discharge at an audio rate and develop the high plate voltage needed to modulate the final. At 5KW the plate voltage is 5KV in some installations a little higher of lower in different cases. to modulate 100% you have to double the plate voltage on the final and bring it to zero for 100% negative modulation so as you can see to get over 100% modulation you need even more HV. The MW5 I was refereing to runs at 4KW and 3800 volts plate voltage. The main HV suply runs 13.5 KV. Now this is a series modulation scheme so with the final at the top of the stack and the modulator at the bottom it can easily control the final plate voltage as plate voltage is what the final sees from cathode to plate and not what is impressed on the plate terminal. Similarly a class A series modulator would have the same voltage requirement as well as current requirements. The difference between the two is that the PDM tube is either on or off so it dissipates less heat. The class A tube has it's resistance changed by grid bias to do the control of the voltage dropped across the modulator. So at part 15 levels the complexity of the circuit to keep the modulator cool is kind of over kill. Also the class A modulator can have better distortion figures before negative feedback is applied so you can do it if you wish but you will need the wave shaping circuits and pulse shaper after modulation is applied to the comparator. The modulator will also need a high reactance at 70KHZ high voltage coil to develop the voltage and the high speed damper diodes to complete the modulator stage. I will try to explain it in more detail if you need. This has been an over view of what is going on there.

Bob


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Fri 24, 2017 11:16 pm 
Member

Joined: May Wed 16, 2012 6:58 pm
Posts: 255
Location: Eastlake, Michigan
Had a MW50, supply voltage 24Kv, PA plate 9.3Kv @ 6 Amp.


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Sat 25, 2017 4:22 am 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 282
Location: pensacola fl
I remember an engineer in Miami Fla back in the 70's installed a MW50. It was an imressive machine but the old RCA Ampliphase across town was huge. Then there was the Monster GE at WABC.


Top
 Profile  
 
 Post subject: Re: Why not a PWM xmtr project?
PostPosted: Nov Mon 27, 2017 12:43 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2073
Location: Saskatoon
Here's a bit more info on the PWM homebrew that I built, as best I can recollect. I never made a schematic, so I'm going from memory. Anyway, there wasn't much to it.

The sample rate was at the carrier frequency, and the pulse width varied in proportion to the audio. I don't recall how I generated the variable pulse width, but there are lots of simple circuits to do this. I think it ran at 800 kHz, because I had a bunch of 800 kHz ceramic resonators at that frequency. The width modulated 800kHz pulses were applied to the gate of a power MOSFET. The source was grounded, and the drain connected via a parallel tank to Vcc. RF was taken off the drain and the tank was supposed to tune the output to the antenna and attenuate harmonics.

Once I had it breadboarded and realized how badly it sounded, I did a bit of mental arithmetic, and realized that this thing wouldn't be generating proper sidebands at all. The pulse width couldn't go above 50% duty cycle, or else the RF output would start to decrease. And the audio would end up being distorted regardless of the modulation level. So that ended the experiment.

I did another one not long after that using Delta-Sigma modulation. Results were equally bad.

This isn't to say that it can't be done. But, it needs more thought put into it than what I did.


Top
 Profile  
 
Post New Topic Post Reply  [ 8 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 10 guests



Search for:
Jump to:  


















Privacy Policy :: Powered by phpBB