user TURBO?s replication of Steven Mark?s TPU ?

[Grumpy]

To all of the EE's here:

Might you be able to squeeze some sort of diode or FET double balanced mixer out of this coil configuration?

[starcruiser]

starcruiser,

Won't the coils create a small impedence bump on the collector?

Varying winding spacing will vary the capacitance and can create a huge impedence.

@Grumpy,

They might if you swept them. I was talking about the appearence of a small ringing signal that shows up after the collector coil is covered with the control coils. It seems that the control coils pickup and excite the collector. This signal, when you  see it, looks like (probably is) the collector ringing from a small induced signal.

Basically what I am saying is the TPU configuration allows self resonance, what we need to do is excite this further to get our power out and/or  feed this signal back to reinforce the self oscillation.

This is shy I stated that we should record the inductance values of the coils as well as the inter-coil capacitance since these will definitely have an effect on the TPU's operation or at least its sweet spot.

Once we know these values we can balance the coils inductance. We could also adjust the inductance values for the 2nd and third collector stages to the harmonics we are looking for to generate the resoance and tap the power.

I agree with X that we should experiement with 3 oscillators and see what results we get.

I will be doing this as soon as I complete my last TPU, I haven't compelted it since I had run out of wire so I had to run out and get the additional wire I needed yesterday. Then of course my daughter had a xmas concert I went to so that provided another delay.

Anyhow I an looking forward to finishing the TPU tonight and building a couple of more oscillators to play with the hetrodyning effect in the TPU. I will share my results afterwards.

[Grumpy]

$#!+  !!!

I'm one FET short - no Schottky's in the drawer (DOH!) 

(edit - fixed typo)

Italian site mentioned that center of TPU looked like a diode bridge - got me thinking...mixer!

Like this (FET's shown):

[PaulLowrance]

A researcher does not spend all day exchanging Internet posts.  A researcher spends most of their time in either deep thought or in the laboratory and *then* every so often exchanges the results of completed experiments and ideas.   God bless you all!

[mrl]

@mramos

I see where you are coming from. I think you would have better control if you could drive the frequencies with your PC. It would probably be a lot easier to keep track of what you've done also. I hope MRL can come up with something for us. What do you think MRL. Could we get a quantity discount from you? There is only a little under 4000 of us

I do not recall him (Mike) stating there was a price. ....

I am with him, waiting on something that looks real and just watching for now.  If someone gets something that is close it will require better timing to keep it running, even if random timing.  And someone to figure out why it is working.  But right now.  Nothing is working and a few are all over it trying.  And I do hope they do.

I am not trying to make wave, just backing Mike up.  "I" believe he is 100% correct, and I might be biased

----------------------
I was thinking about the design last night.

The unit would work on an RS485 network.  You can get USB to RS485 boxes for about 45$ US.  RS485 is very inexpensive (and robust) and you can string 32 nodes on one segment without having to use a repeater.

The system would use a PIC micro controller (worth about $9 US) and some sort of frequency synthesizer.  The MAX 038 would be a good choice but Maxim does not recommend it for new designs due to it being manufactured by someone else and they are not working with that wafer process anymore (I've emailed Maxim about this).

The PIC micro controller counter timer would be set up as a frequency counter which would read the synthesizer's output frequency.  It would then adjust and maintain the synth output via an inexpensive high res DAC.  You could also control the duty cycle too.  There would only be one synthesizer and controller per board.  You would string more boards on the 485 buss network if you wanted more frequencies.  For three frequencies you would used three boards.   You could attach any drivers you want to the board output.  Maybe the board would have prevision for soldering a driver IC and FET / IGBT to it for 1 ? 2 Amp square wave switching.  The board could also have a frequency counter input on it for external measurements (TTL / CMOS level only).  The board would measure about 2.5 x 2.5 inches or so.  The cost would be about $150 US in QTY 1 (with Windows Software) and $100 QTY for 2 and about 70 each for three and up.  I could cram other features on it as well such as a digitally controllable power supply as long is it didn't drive the cost up to an unattractive level

If people showed an interest in these lab tools I could also design other 485 network add-ons such as high speed power sequencers for pulsing coils.  You would use Windows software to precisely program the sequencer frequencies, timing and duty cycle -- and possibly even the voltage output.  These boards would be affordable.

Before I would go a head with the design I would have to see some major interest in it.

Prices subject to change with notice.

Mike