Hi, I am looking to build an amplifier which amplifies mainly highish frequencies (50Khz-13MHz) and I decided upon a class D amplifier because efficiency is very important. High voltage up to ~1000V is required and very low current is needed of about a microamp. After the amplifier i decided to use a cockroft walton stack to pump up the voltage.
All was looking fine until I had a browse in an electronics book and I can't really figure why a specific amplifier type (Class D or any other class) would be chosen over a simple common emitter amplifier. I suspect the reasons are obvious but they aren't jumping out at me at the moment!
Thanks for any help
All was looking fine until I had a browse in an electronics book and I can't really figure why a specific amplifier type (Class D or any other class) would be chosen over a simple common emitter amplifier. I suspect the reasons are obvious but they aren't jumping out at me at the moment!
Thanks for any help
You haven't really said what you are trying to do.
50khz to 13mhz is a wide range (LRF to RF) to amplify. And you want 1000 volts output !!!
What is the input to this amplifier ?
A CW multiplier gives a high voltage output from a steady ac input... so I don't really see the connection between an amplifier and this.
Forget amplifier classes at the moment... what are you trying to achieve ?
50khz to 13mhz is a wide range (LRF to RF) to amplify. And you want 1000 volts output !!!
What is the input to this amplifier ?
A CW multiplier gives a high voltage output from a steady ac input... so I don't really see the connection between an amplifier and this.
Forget amplifier classes at the moment... what are you trying to achieve ?
The amplifier is for a university project and I need to transmit an oscillating signal (which ideally should be 13.56MHz but component limitations mean we are starting at around 100K and if that works we'll stick with it). The oscillating signal is not going to vary so I only need to amplify one frequency, and the high voltage is required because the amp must be capacitively coupled across a physical air gap between two copper plates.
So the input is 100K sine wave for now of about 10V
What do you think would be best for this? The device should ideally be battery powered which is why I wanted a class D amp. Despite the high voltage I thought batteries would be ok because the current drawn here is tiny. The CW stack should be ok I think because of the constant signal.
So the input is 100K sine wave for now of about 10V
What do you think would be best for this? The device should ideally be battery powered which is why I wanted a class D amp. Despite the high voltage I thought batteries would be ok because the current drawn here is tiny. The CW stack should be ok I think because of the constant signal.
Hmmm, I'm something familiar with broadband RF amplification and what you want here ain't gonna happen in one box without bandswitching. Even broadband military rigs will have at least three power amplification modules to cover this range.
There's some projects around qsl.net that can do class-D broadband nicely with a pair of IRF510's and you could use toroidal step-up transformers (use good teflon wire) to get your voltage.
Cheers!
There's some projects around qsl.net that can do class-D broadband nicely with a pair of IRF510's and you could use toroidal step-up transformers (use good teflon wire) to get your voltage.
Cheers!
Class D amps not being fast enough was a bit of a worry but there are some pretty fast components out there and i've seen amps capable of 10MHz and I found mosfets with some good charactoristics so it doesnt seem too much of a problem acheiving the frequency needed. Besides, as I said I hope the device to work at lower frequencies.
Geek - I think you misunderstood the design in that I amp only amplifying 1 frequency. I probably wasn't clear but I want the design to work at around 100K but if I am struggling to transmit across an air gap I will increase the frequency so help matters. Using a high voltage was there to allow me to reduce frequency. High frequency is problematic in all parts of the design so I don't really want to use it if at all possible!
So yes for the time being its best off said that I'm designing an amplifier to significantly increase the voltage a 100KHz sine wave. I'll look into the torroidal transformers to see whats available. Do you think it is possible to run such a device off batteries. The idea was for a hand held device you see. It is possible to create ~30KV using a 9V battery and a flyback transformer so I thought all hope wasn't lost!
Geek - I think you misunderstood the design in that I amp only amplifying 1 frequency. I probably wasn't clear but I want the design to work at around 100K but if I am struggling to transmit across an air gap I will increase the frequency so help matters. Using a high voltage was there to allow me to reduce frequency. High frequency is problematic in all parts of the design so I don't really want to use it if at all possible!
So yes for the time being its best off said that I'm designing an amplifier to significantly increase the voltage a 100KHz sine wave. I'll look into the torroidal transformers to see whats available. Do you think it is possible to run such a device off batteries. The idea was for a hand held device you see. It is possible to create ~30KV using a 9V battery and a flyback transformer so I thought all hope wasn't lost!
OK then.... ha what do I know
Still not 100% clear on this. If I'm talking rubbish please tell me... I just don't get what you are trying to do.
A CW stack generates a high DC voltage from an AC one... and you can't "transmit" DC... it doesn't propagate as a EM wave. The CW stack and it's diodes mean the input frequencies are "low", perhaps up to a few khz at most (I would guess... because you need high speed diodes for HF, to eliminate losses).
You can generate something that will "jump" from one electrode to another but that's not transmitting. The fundamental frequency you put into the CW is "lost" as it just becomes DC.
Still not 100% clear on this. If I'm talking rubbish please tell me... I just don't get what you are trying to do.
A CW stack generates a high DC voltage from an AC one... and you can't "transmit" DC... it doesn't propagate as a EM wave. The CW stack and it's diodes mean the input frequencies are "low", perhaps up to a few khz at most (I would guess... because you need high speed diodes for HF, to eliminate losses).
You can generate something that will "jump" from one electrode to another but that's not transmitting. The fundamental frequency you put into the CW is "lost" as it just becomes DC.
Is it possible to generate 30kv from 9 volts... yes of course.
Just clarify something... are you trying to recover the fundamental (100khz/ 13mhz whatever across this air gap or are you just trying to generate a voltage to cause the air to breakdown and for a spark to jump across ?
Are you trying to transmit "information" across this gap by modulating the carrier in some way ?
Just clarify something... are you trying to recover the fundamental (100khz/ 13mhz whatever across this air gap or are you just trying to generate a voltage to cause the air to breakdown and for a spark to jump across ?
Are you trying to transmit "information" across this gap by modulating the carrier in some way ?
I've never used a CW stack before so you're points are no doubt very relevant. I see the problem you are saying - I was unaware that the stack produced a DC output and if that's the case then it is indeed about as useful as a chocolate teapot! Might be best off with a transformer or something else. Any ideas?
Lol... two copper plates separated by air form a capacitor.
If you apply an hf signal to one, the signal will "transfer" to the other. The higher the frequency the "more" signal you get in simple terms.
So a low voltage amp would recover that... even opamps at 100khz would work. It would be prone to pickup of 50hz mains interference but that's easily solved. The signal although small would be easily recoverable.
If you just connect an osciloscope probe to one plate and the other to a signal generator you will see the signal.
Is that all you are trying to do ?
If you apply an hf signal to one, the signal will "transfer" to the other. The higher the frequency the "more" signal you get in simple terms.
So a low voltage amp would recover that... even opamps at 100khz would work. It would be prone to pickup of 50hz mains interference but that's easily solved. The signal although small would be easily recoverable.
If you just connect an osciloscope probe to one plate and the other to a signal generator you will see the signal.
Is that all you are trying to do ?
You can even do this even at audio frequencies... just "break" the signal path between the output of say an opamp and the input to another and you will still "hear" the audio. It's capacitively coupled across the break. If you connected two plates there, one as "transmitter" and one as "receiver" then the effect would be clear. It would sound "tinny" because the capacitor would only be a few 10's of picofarads and that would limit low frequencies.
If you designed the the "receiving amplifier with a very high input impedance then that effect would be less.
If you designed the the "receiving amplifier with a very high input impedance then that effect would be less.
That's pretty much what I need to do yes. The device on the other side of the plates is passive but the capacitive coupling should still work. The gap needs to be about 1cm and 30V is needed on the passive side. There's no way to amplify the signal on the receiving side which is why I wanted to send it using a high voltage. I'm a little bit confused by the requirements now! What sort of voltage do you think would be necessary using a 1cm square plate and having a 1 cm gap. We did a capacitance measurement of a smaller gap and it was in the picofarad range.
It's probably best for you to draw out as a block diagram what you want, and put in signal sources and levels.
What do mean by "the device on the other side is passive" ? but needs 30 volts ? don't understand Passive... no power... why 30 volts. Do you mean you need 30 volts of signal coupled through the gap ?
What are you going to do with the signal on the receiving side, just measure it in some way ? how ?
What do mean by "the device on the other side is passive" ? but needs 30 volts ? don't understand
Passive... no power... why 30 volts. Do you mean you need 30 volts of signal coupled through the gap ? What are you going to do with the signal on the receiving side, just measure it in some way ? how ?
I've uploaded a block diagram of the main parts of the design. The "passive" device evidently isn't passive - it's not my part of the design but i think it was called passive due to the lack of battery but it it evidently isn't as it needs 30V to wake it up! The signal is rectified on the other side to power the components. Can't really say whats happening on that side due to NDA.
Attachments
The parts to the left in your diagram are pretty straightfoward.
It's what's on the right that's not. You can get the "signal" onto the receiving side, but to me the 30 volts 1ua bit doesn't mean anything.
There are a lot of practical problems. Whatever you use to measure or receive that signal is going to have to be powered (active).
If you apply the signal to the left plate (say 100khz sine) then it should be recovereable on the other side, but only with some carefully thought out electronics... perhaps an opamp using guard rails around the inputs to maximise the input impedance although that is more for DC measurement.
http://www.national.com/an/AN/AN-241.pdf
Lol... it's some weird project... what it's all for ?
It's what's on the right that's not. You can get the "signal" onto the receiving side, but to me the 30 volts 1ua bit doesn't mean anything.
There are a lot of practical problems. Whatever you use to measure or receive that signal is going to have to be powered (active).
If you apply the signal to the left plate (say 100khz sine) then it should be recovereable on the other side, but only with some carefully thought out electronics... perhaps an opamp using guard rails around the inputs to maximise the input impedance although that is more for DC measurement.
http://www.national.com/an/AN/AN-241.pdf
Lol... it's some weird project... what it's all for ?
Thanks for the help The NDA states that I can't give details of the project so can't really say what its all for which is a bit unhelpful unfortunately! The device on the right is already designed elsewhere and the monitoring of the transmitted signal is sorted however i will have to match the impedance's to avoid reflections in the transmission. For now I will just look at a suitable way of boosting the voltage.
The NDA states that I can't give details of the project so can't really say what its all for which is a bit unhelpful unfortunately! The device on the right is already designed elsewhere and the monitoring of the transmitted signal is sorted however i will have to match the impedance's to avoid reflections in the transmission. For now I will just look at a suitable way of boosting the voltage.
Oooops!
I read into this you needed 50KHz to 13MHz @ 1KV output across the board.
Cheers!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.