The TPU uncovered? (A PROBABLE technique.)

[duff]

@ Duff,

you're probably the only one interested in this, so i'm just posting the netlist. it's the simple single inductor version, which won't give the exact results as Paul's but close enough. you'll see the differentiation effect and voltage peak anyway:

Code Select Expand


Vgen 1 0 Pulse -10 10 0 10n 10n 6.25u 12.5u
Rgen 1 2 50
L1 2 3 20uH
R1 3 0 1.2


@poynt

Ok - maybe I can learn something here, so please give me some explanation.

To me the circuit does not represent Pauls circuit in that you don't show both bifilar inductors with opposing polarity.

The differentiation is bascially voltage across the inductor as a  result of the RL time constant.

I like the output waveform but I don't understand why the -10 volt pulse

If I apply a positve pulse everything goes to hell - I don't get that at all...


-Duff

[pauldude000]

@duff

He had to do that in part to account for the 20v demonstrated on the screen shot. The scenario would never show 20v, unless 10v+ AND 10v- were input, which my FG was not doing.

My function generator only puts out 10v total amplitude. That is... 10v peak to peak.... (5v+ and 5v- at zero dc offset which I used for the experiment. It can only put out 10v+ only at full dc offset+, and then 0v-. No matter how you look at it, 10v is still unaccounted for.)

http://www.victorelectronics.com/specifications/vc2002_specs.htm

@all

I wish to point out some things which seem to have gotten lost since I first posted:

1. The effect is common. No kidding it has a "common buzzword" in EE. I stated it would.
2. I stated that I was going to have to deal with just this sort of BS, even with the complex test.

Yes, you are looking at "wave differentiation". I still think it is a cute word, with little meaning.

DEFINITION:

http://www.merriam-webster.com/dictionary/differentiation

Look it up. You will quickly find it is a buzzword. The usage as Grumpster posted is used in the manner of a differential op-amp.

Interesting enough, you see this effect to some degree all the time, even in full waveforms.

A non-integrating wave. Just because an effect has a common term applied to it, does not change anything. I explained this was so earlier. Nothing is therefore "relegated away", or "crap" as one so "quaintly" put it.

Paul Andrulis

[poynt99]

@poynt

Ok - maybe I can learn something here, so please give me some explanation.

To me the circuit does not represent Pauls circuit in that you don't show both bifilar inductors with opposing polarity.

The differentiation is bascially voltage across the inductor as a  result of the RL time constant.

I like the output waveform but I don't understand why the -10 volt pulse

If I apply a positve pulse everything goes to hell - I don't get that at all...


-Duff

@ Duff

the single inductor shows you the same effect, just a slightly different wave form. but here is the dual bucking version since you asked. i threw in a small parallel capacitor for good measure.

schematic, wave form, wave form vertical zoomed, more or less like Paul did in his post. the netlist is here also. sorry for the funny net numbers.

i am sure Paul's generator is swinging plus and minus 10V, so that's what i did. you won't get the right levels measured if you use only +10V and 0V. does it make sense now?

[Grumpy]

@all

I wish to point out some things which seem to have gotten lost since I first posted:

1. The effect is common. No kidding it has a "common buzzword" in EE. I stated it would.
2. I stated that I was going to have to deal with just this sort of BS, even with the complex test.

Yes, you are looking at "wave differentiation". I still think it is a cute word, with little meaning.

DEFINITION:

http://www.merriam-webster.com/dictionary/differentiation

Look it up. You will quickly find it is a buzzword. The usage as Grumpster posted is used in the manner of a differential op-amp.

Interesting enough, you see this effect to some degree all the time, even in full waveforms.

A non-integrating wave. Just because an effect has a common term applied to it, does not change anything. I explained this was so earlier. Nothing is therefore "relegated away", or "crap" as one so "quaintly" put it.

Paul Andrulis

Dude,

Differentiation has nothing to do with op amps - it's a form of filtering - aka "high pass filter" - no rocket science and no OU.

Look up "Dirac's Delta Function" and you might learn something - or not...ROFLMFAO-LOL!

No matter how you deny it, how you slice it, dice it, ignor it, or try to explain it away - your "kicks" are just "ordinary pulses".

You can continue in a world of denial or you can try to use these pulses.  I already told you how.

If I am so "off" - then put a sine wave in instead of a square and you will find that sines are relatively "immune" to differentiation - you'll get attenuation and phase shift but it still looks like a sine - go ahead punk! - make my freakin' day!

G

[duff]

@ Duff

the single inductor shows you the same effect, just a slightly different wave form. but here is the dual bucking version since you asked. i threw in a small parallel capacitor for good measure.

schematic, wave form, wave form vertical zoomed, more or less like Paul did in his post. the netlist is here also. sorry for the funny net numbers.

i am sure Paul's generator is swinging plus and minus 10V, so that's what i did. you won't get the right levels measured if you use only +10V and 0V. does it make sense now?

@Ponyt,

I understand what your saying however to me it does not account for the way Paul
made the connections. I'm still thinking  my orginal post is closer even though
I did not get the correct voltage levels.

Your bucking coils I buy, but they are not bifilar and reflect the connection Paul
made.

The schematics will hopefully illustrate what I'm saying.




I appreciate your work in posting the models and hope you can see my point.



-Duff