Understanding electricity in the TPU.

[e2matrix]

@All

I put up a new short video located here;
http://www.youtube.com/watch?v=vUH4uIbvRBo

@gyulasun

Thanks a heep for coaching me through this. I see now the error in my ways. I won't promise not to make more errors even intentionally (just to see the effects - lol). One of my breadboards looks like a cratered moon surface so I am used to it.

I thank you for your kind patience.

@otto

I answered you in @GK's thread.

You've probably already seen this but it sounds like what GK just mentioned in other thread may explain why your voltage went up in your last video when you disconnected the ground lead on the FG.  It does sound like results will be much better if everything is isolated from any grounds which all normal equipment will have if plugged in the wall. 

[gyulasun]

@All

I put up a new short video located here;
http://www.youtube.com/watch?v=vUH4uIbvRBo

@gyulasun

Thanks a heep for coaching me through this. I see now the error in my ways. I won't promise not to make more errors even intentionally (just to see the effects - lol). One of my breadboards looks like a cratered moon surface so I am used to it.

I thank you for your kind patience.

@otto

I answered you in @GK's thread.

Hi Wattsup,

I have just watched your 2nd part video and you demonstrated how bad a piece of advice I gave to you when I was suggesting to connect the FG's negative output to the negative pole of the power supply: I deprived you from several tens of Volts....  but fortunately you realised this by disconnecting the negatives and regained the lost voltage.   

I cannot give a strait explanation for this "loss"  maybe the lack of the connection somehow modifies the falling edge of the pulse and makes it falling quicker, a qiucker switch-off can explain the increasing regained voltage from the collapsing field. (The quicker you switch the current off, the higher the pulse amplitude becomes.)  Now, why the switch-off improves? good question...

One thing is sure: in pulse technique the shorter cables, wires are used the better and dependable the results are.  I mean the too long wires leading to and from the gate current meter to the gates, this is one thing. Another thing is I cannot figure out what discharges the gate-source capacitances when the gates have no direct closed DC path via the the FG's 50 Ohm output when you disconnect the negatives.
You are very lucky having that HP FG, it has a very sharp rising and falling edges and high output amplitude, this is a big blessing for you because if you were to use the bipolar type NE555 timer IC for generating the pulses, you would be settled with some hundred nanosecond rising and falling times (NB: the CMOS LMC555 or TLC555 has a much better rising and falling times at their output than the bipolar version has).  Then you would be forced to use high speed MOSFET drivers Otto referred to between the 555 output and the power MOSFET switch gate (negative point of the 555 and the n channel MOSFET source electrode are to be connected  ).

So all I wish keep up your excellent and steady work and practice makes perfect.

Hopefully Otto will not find this a mess... ("What a mess" he wrote after my previous long ramblings  )

rgds,  Gyula

EDIT Just read GK's thoughts on using totally ground independent setups, I agree this can explain your voltage rise too (maybe the scope you used between the drain-source electrodes insured a better negative connection and using a second negative connection already created a ground-loop with the existing one).

[giantkiller]

I have taken Otto's responses like that means 'just the facts, minimize, focus with just the pertainent, reject the noise'.

I have seen this in myself and most others here and that is if you have 5 things to focus on the focus usually goes to the apparent and not the hidden or obscure. And down the rabbit hole we go. This is one of Grumpy's pet peaves. How quickly we get lost...

Even though wading through the boastings and postings takes time it still feeds the knowledge.

@wattsup,
I have built a 3 freq 555 based timer box with a potentiometer for timing and width. Then put the timing capacitor connections outside the project box for 3 Elenco capacitor selection boxes hookup.
I'll take a pic and posted tonight.
I also have a box based on 3 xr2206's for square, triangle and sine waves. Both of these drive fet gates.
Also using a fet to drive a fet would enable the second fet to slam change really sharp because ofd the hard on and off transistion.

[gyulasun]

Thanks GK,  I did not really mean it...   

[giantkiller]

You lost me on this one. No problem though.

Thanks GK,  I did not really mean it...   

All have given input over time to get us at this point to nail this down. Now we have minimal amount of patterns, facts, patents, people and threads. And that equates to speed.