QCW Documented
Goodchild, Sat Nov 19 2011, 09:51PMHello every one, there has been a lot of talk about QCW lately on the forum and lots of question being asked.
I decided to do a nice well documented page on my experience with the QCW on my new web site. As far as I know I was the first to duplicate Steve Ward's results (This is unconfirmed). with the class D bus voltage modultion
Enjoy:
Fell free to ask question.
Re: QCW Documented
lightlinked, Tue Nov 22 2011, 06:32AM
Thank you for doing this. Sharing and documenting our research is what makes us a scientific community.
Are you using a PFC boost converter or a rectifier?
lightlinked, Tue Nov 22 2011, 06:32AM
Thank you for doing this. Sharing and documenting our research is what makes us a scientific community.
Are you using a PFC boost converter or a rectifier?
Re: QCW Documented
Avalanche, Tue Nov 22 2011, 01:16PM
I finally got around to reading it - interesting article, thanks
I'm looking forward to seeing how things develop in the QCW world, even if I never get around to trying anything properly myself. The audio modulation should be particularly entertaining, if anyone can find any IGBTs that can cope with the abuse
Avalanche, Tue Nov 22 2011, 01:16PM
I finally got around to reading it - interesting article, thanks
I'm looking forward to seeing how things develop in the QCW world, even if I never get around to trying anything properly myself. The audio modulation should be particularly entertaining, if anyone can find any IGBTs that can cope with the abuse
Re: QCW Documented
Dr. Drone, Tue Nov 22 2011, 03:34PM
Dr. Drone, Tue Nov 22 2011, 03:34PM
Re: QCW Documented
Steve Conner, Tue Nov 22 2011, 06:04PM
Dr. Spark, you should modulate the plate voltage on your Quadzilla. You could call it QuasiModZilla.
Steve Conner, Tue Nov 22 2011, 06:04PM
Dr. Spark, you should modulate the plate voltage on your Quadzilla. You could call it QuasiModZilla.
Re: QCW Documented
Goodchild, Tue Nov 22 2011, 06:30PM
Thanks for the positive feedback. To answer a couple of question.
No I don't use PFC "yet". I have yet to decided if it would really be worth it to add an extra layer of complexity to this project. Because I can see it know, one 60N60 in the full bridge fails, this shorts the other 60N60 in the half bridge, this puts a dead short on the modulator, shorting all it's 10mF of bus C and kill it, this then puts a short on the PFC and possibility killing it! It's a scary thought, a big domino effect of dead silicon
Audio modulation has already been tried on the QCW by Steve Ward, and the results of very cool! It's just a matter of using only half the audio signal to maintain <50% duty operation.
Man you are going to need one hell of a modulator to do QuadZilla!!!
Goodchild, Tue Nov 22 2011, 06:30PM
Thanks for the positive feedback. To answer a couple of question.
No I don't use PFC "yet". I have yet to decided if it would really be worth it to add an extra layer of complexity to this project. Because I can see it know, one 60N60 in the full bridge fails, this shorts the other 60N60 in the half bridge, this puts a dead short on the modulator, shorting all it's 10mF of bus C and kill it, this then puts a short on the PFC and possibility killing it! It's a scary thought, a big domino effect of dead silicon
Audio modulation has already been tried on the QCW by Steve Ward, and the results of very cool! It's just a matter of using only half the audio signal to maintain <50% duty operation.
Man you are going to need one hell of a modulator to do QuadZilla!!!
Re: QCW Documented
Gregory, Tue Nov 22 2011, 07:25PM
QuadZilla with a modulator will be a beautiful setup.
Eric,
Maybe you can try some OCD at the buck inverter output, before the output LC filter. One 100A shunt can do the job!
Gregory, Tue Nov 22 2011, 07:25PM
QuadZilla with a modulator will be a beautiful setup.
Eric,
Maybe you can try some OCD at the buck inverter output, before the output LC filter. One 100A shunt can do the job!
Re: QCW Documented
Steve Conner, Tue Nov 22 2011, 09:13PM
If Phillip Slawinski's "SSVC" can do this by modulating one 833
Just think what Quadzilla could do with four of them. It would be monstrous, you must do it Dr. Spark!
The QCW and the SSVC are pretty much identical, and the same theory applies.
Steve Conner, Tue Nov 22 2011, 09:13PM
If Phillip Slawinski's "SSVC" can do this by modulating one 833
Just think what Quadzilla could do with four of them. It would be monstrous, you must do it Dr. Spark!
The QCW and the SSVC are pretty much identical, and the same theory applies.
Re: QCW Documented
Goodchild, Tue Nov 22 2011, 10:02PM
Yes I was originally planning on having a LEM100 DC current transducer do the job. But I never got around to adding it. I will most likely implement this in the final design.
Goodchild, Tue Nov 22 2011, 10:02PM
Gregory wrote ...
QuadZilla with a modulator will be a beautiful setup.
Eric,
Maybe you can try some OCD at the buck inverter output, before the output LC filter. One 100A shunt can do the job!
QuadZilla with a modulator will be a beautiful setup.
Eric,
Maybe you can try some OCD at the buck inverter output, before the output LC filter. One 100A shunt can do the job!
Yes I was originally planning on having a LEM100 DC current transducer do the job. But I never got around to adding it. I will most likely implement this in the final design.
Re: QCW Documented
Dr. Dark Current, Tue Nov 22 2011, 11:18PM
Steve, AFAIK Phillip used a ridiculously high voltage on the plate which surely doesn't help the tubes life. Maybe if you used such a high voltage transformer with a standard doubler and a staccato, you could get similar results...
Dr. Dark Current, Tue Nov 22 2011, 11:18PM
Steve, AFAIK Phillip used a ridiculously high voltage on the plate which surely doesn't help the tubes life. Maybe if you used such a high voltage transformer with a standard doubler and a staccato, you could get similar results...
Re: QCW Documented
Dr. Drone, Wed Nov 23 2011, 01:40AM
Dr. Drone, Wed Nov 23 2011, 01:40AM
Re: QCW Documented
Goodchild, Wed Nov 23 2011, 02:21AM
Hey Dr. Spark,
I bet you could build a modulator for quadZilla that wasn't supper big. Using a buck or SLR topology.
You could probably even do 4 individual supplies to drive the 4 plates! You could force current sharing among the tube this way even with them unmatched!
It would be supper cool to see a supply do this, I can only imagine qaudZilla with AM audio modulation. 50Kw plasma sub woofer hehe
Goodchild, Wed Nov 23 2011, 02:21AM
Hey Dr. Spark,
I bet you could build a modulator for quadZilla that wasn't supper big. Using a buck or SLR topology.
You could probably even do 4 individual supplies to drive the 4 plates! You could force current sharing among the tube this way even with them unmatched!
It would be supper cool to see a supply do this, I can only imagine qaudZilla with AM audio modulation. 50Kw plasma sub woofer hehe
Re: QCW Documented
teravolt, Wed Nov 23 2011, 02:46AM
Would it be pausible to just modulate the grid of a 833 to create a ramped power envilope? This could be done by using a PWM to tern the tube on for a spesific time period that would be equal to 90 deg or less of the primary tank cycle. Some how the ZCS would have to syncronise with the PWM rising edge so the tube turns on at the right time. The tube is a natural choice to excite a parallel primary tank cycle.
teravolt, Wed Nov 23 2011, 02:46AM
Would it be pausible to just modulate the grid of a 833 to create a ramped power envilope? This could be done by using a PWM to tern the tube on for a spesific time period that would be equal to 90 deg or less of the primary tank cycle. Some how the ZCS would have to syncronise with the PWM rising edge so the tube turns on at the right time. The tube is a natural choice to excite a parallel primary tank cycle.
Re: QCW Documented
Marko, Wed Nov 23 2011, 03:48AM
Hi guys
Well my personal evil plan was to use two pulse tubes in a half bridge configuration to base drive a resonator directly ;)
Depending on whether I could beat Steve in making the buck-less QCW work there might be a number of ways to achieve QCW operation with such a setup (including feeding a ramp to screen grid if tubes used are tetrodes)
Wondered if those GMI6 tubes would be good for start...
But too many a project already so better keep this dreamware for now...
Great work guys!
Marko
Marko, Wed Nov 23 2011, 03:48AM
Hi guys
Well my personal evil plan was to use two pulse tubes in a half bridge configuration to base drive a resonator directly ;)
Depending on whether I could beat Steve in making the buck-less QCW work there might be a number of ways to achieve QCW operation with such a setup (including feeding a ramp to screen grid if tubes used are tetrodes)
Wondered if those GMI6 tubes would be good for start...
But too many a project already so better keep this dreamware for now...
Great work guys!
Marko
Re: QCW Documented
Goodchild, Wed Nov 23 2011, 06:43AM
I imagine if you did some kinda grid modulation it would be similar to Steve's delta sigma modulation. As you would be forced to switched at ZCS or with ZVS. So the low freq ripple would be present on the output.
Goodchild, Wed Nov 23 2011, 06:43AM
I imagine if you did some kinda grid modulation it would be similar to Steve's delta sigma modulation. As you would be forced to switched at ZCS or with ZVS. So the low freq ripple would be present on the output.
Re: QCW Documented
Luceš, Sun Nov 27 2011, 04:55AM
So all Phillip did was feed the plates a ramp current? Just with an inverter? Just curios...
Luceš, Sun Nov 27 2011, 04:55AM
So all Phillip did was feed the plates a ramp current? Just with an inverter? Just curios...
Re: QCW Documented
Marko, Sun Nov 27 2011, 01:25PM
Nope, he used a modulated SLR power supply with a big IGBT brick, as far as I recall. Which I guess is about the same hassle as building a buck converter for a DRSSTC.
I think the first and easiest approach for tubes would be to modulate the tetrode screens. I think it might be possible to use a big inductor in series with power supply to ease the losses on the tube.
Next would probably be sigma-delta QCW. One could use "staccato" modulator to turn the tube on and off while sensing the primary current. So far it's unknown whether this would work.
Finally, there's the possibility of using two big high voltage pulse tubes to feed the resonator base directly. If I ever get a chance to own some awesome pulsed tubes this is how I would build a QCW.. I also think sigma delta would have far more chance of working that way due to 1-order system and naturally slower energy transfer time constant.
Marko
Marko, Sun Nov 27 2011, 01:25PM
crazyivan wrote ...
So all Phillip did was feed the plates a ramp current? Just with an inverter? Just curios...
So all Phillip did was feed the plates a ramp current? Just with an inverter? Just curios...
Nope, he used a modulated SLR power supply with a big IGBT brick, as far as I recall. Which I guess is about the same hassle as building a buck converter for a DRSSTC.
I think the first and easiest approach for tubes would be to modulate the tetrode screens. I think it might be possible to use a big inductor in series with power supply to ease the losses on the tube.
Next would probably be sigma-delta QCW. One could use "staccato" modulator to turn the tube on and off while sensing the primary current. So far it's unknown whether this would work.
Finally, there's the possibility of using two big high voltage pulse tubes to feed the resonator base directly. If I ever get a chance to own some awesome pulsed tubes this is how I would build a QCW.. I also think sigma delta would have far more chance of working that way due to 1-order system and naturally slower energy transfer time constant.
Marko
Re: QCW Documented
Goodchild, Sun Nov 27 2011, 07:48PM
SLR and buck are both noteworthy systems that would work to modulate a coil.
SLR has the advantage of being more efficient because of ZCS and also inherent current limiting. This come at the price of complexity though.
Buck also works but is less efficient and needs a separate current limiting circuit, but they are darn easy to get working and have very few parts.
In the end it's up to preference. Although I like the idea of no modulator and doing all the modulation on the bridge or with the tube screen, I still have doubts about it's piratical application. Mainly the ripple present in this kind of system and also the lower coupling required for it to work properly. But I'm still hopeful!!!
Goodchild, Sun Nov 27 2011, 07:48PM
SLR and buck are both noteworthy systems that would work to modulate a coil.
SLR has the advantage of being more efficient because of ZCS and also inherent current limiting. This come at the price of complexity though.
Buck also works but is less efficient and needs a separate current limiting circuit, but they are darn easy to get working and have very few parts.
In the end it's up to preference. Although I like the idea of no modulator and doing all the modulation on the bridge or with the tube screen, I still have doubts about it's piratical application. Mainly the ripple present in this kind of system and also the lower coupling required for it to work properly. But I'm still hopeful!!!
Print this page