HFU HF Underground
Technical Topics => The RF Workbench => Topic started by: OgreVorbis on November 10, 2018, 0744 UTC
-
This may sound like a stupid question to those who know more about RF engineering, but...
Why have I never seen an RF MOSFET PWM modulated or other high level modulated?
They are always used as linear amplifiers.
I have a BLF188XR amplifier board and I was wondering if it is possible to modulate it this way? Is low level modulation the only option with this?
Why does every PWM use GaN or SiC or those IRFP (whatever they're called)?
-
No it won't work as its a LINEAR AMP.
As soon as you reduce the voltage it will cease to function!
You need Class E which is way more efficient, simpler, cheaper etc etc.
They use those devices you quoted as they are modern, i.e. available, are cheap and are very easy to drive and have a low on resistance. There's no need to use GaN as they're for much higher (RF) frequencies. No real need to use SiC either unless the voltages are very high. Any crappy ol' FET will work as a PWM output as the frequency is only 200KHz or so....you just DONT want Ohms of on resistance as with highish current flow you'll drop volts and generate a lot of heat. SiC FETs can be use with high voltages keeping the current low, I2R is what generates loss.
With a linear amp you're best directly modulating the DDS and amplifying that to 3W to drive the linear. If you check my old posts I mention directly modding a AD9850 and there's a good article on the web about such..
Bought the same amp as you to do said build! You may beat me to it! Was also going to make a 3W SSB generator to drive it too!
Keep us posted!
Str.
-
No it won't work as its a LINEAR AMP.
As soon as you reduce the voltage it will cease to function!
You need Class E which is way more efficient, simpler, cheaper etc etc.
They use those devices you quoted as they are modern, i.e. available, are cheap and are very easy to drive and have a low on resistance. There's no need to use GaN as they're for much higher (RF) frequencies. No real need to use SiC either unless the voltages are very high. Any crappy ol' FET will work as a PWM output as the frequency is only 200KHz or so....you just DONT want Ohms of on resistance as with highish current flow you'll drop volts and generate a lot of heat. SiC FETs can be use with high voltages keeping the current low, I2R is what generates loss.
With a linear amp you're best directly modulating the DDS and amplifying that to 3W to drive the linear. If you check my old posts I mention directly modding a AD9850 and there's a good article on the web about such..
Bought the same amp as you to do said build! You may beat me to it! Was also going to make a 3W SSB generator to drive it too!
Keep us posted!
Str.
Thank you for the response, and if you don't mind, see my other post in your thread about your 400W PWM.
What I meant to say about the BLF188 is: Why don't they use it in Class E configuration? Is it not possible? Is it the on resistance you where talking about?
-
BTW: I have a 20W PEP amp with 2x RD15HVF1 and the BLF188 (which I haven't finished constructing yet). I plan to drive the BLF with the former at reduced power.
I am using the 9850 DDS board I got from Germany on eBay (forgot the name). It has AM modulation built in. I think it is rf2017.
I will test it out soon, but I need a water cooling block and I'm trying to decide if I should buy or build one. I think I'll build it because I spent a lot of time looking for a good one. Not just size wise, but also with known thickness and tube pattern, so I know I am not going to puncture it when I drill the holes to mount the BLF. I'd rather not used forced air. It will just be bulky and loud. I already have a good radiator which should be more than capable.
-
BLF188's and their ilk are designed primarily for linear and class B service. For the amount of power you get at HF for AM anyway it is far cheaper to use standard power conversion type fets to generate the high voltage swings necessary to get decent power, unless you are using something other than AM.
For the record, I've found SiC fets nearly ideal for RF generation at HF frequencies because they are easy to drive and once you get to the 300-500W carrier level the voltages the fets must endure enters the realm of SiC's. It should be possible to make 2.5KW carrier out of as little as two SiC devices, however water cooling would be required. With 4 devices, air cooling is practical again. One could combine 4 of these modules and achieve 10KW carrier levels with overhead for combining losses and asymmetrical modulation.
Maybe some day...
+-RH
-
Max volts on the BLR is 50ish.
Class E generates 3.5 X Vcc on drain so you could ONLY use 50/3.5=14V2, not much use to generate any real power!
So SiC it is!
CREE.
Str