STEAP and the TPU

[Centraflow]

Thanks for that link ,I had located that already and as far as i can see all is good .

For the basics how about the cap values for the 3 4047's and the 40106's .

I cant recall where i got them from in the first place but they are not in this thread yet.
It may be that things are ok but i have a timing problem

Use this calculator for your RC frequency (remember you need twice the frequency because the Q outputs are halved.

Regards

Mike

[Centraflow]

oops forgot the link

https://www.omnicalculator.com/physics/rc-circuit



For the duty, you can use 10K and 560pF, but it is just an RC timing, so use the RC timing app in the same place.

Regards

Mike

[Centraflow]

Lindsay, make sure you have adjusted the duty of "C" before the 4011 sync IC.

See attached

Regards

Mike

[3D Magnetics]

Perhaps a cro picture ? I have a ,c, and b  on together most of the time
Astable c controls all outputs frequency and phase  in sync.
Im not able to get b inhibited by c as it should right?

All pulse shaping works ok .

I could have fake 4047's ? highly unlikely
what values did you use for the 4047 rc?

Its the same on both boards which almost rules out operator problem

Dave ...did you have any issues ?

[Centraflow]

Perhaps a cro picture ? I have a ,c, and b  on together most of the time
Astable c controls all outputs frequency and phase  in sync.
Im not able to get b inhibited by c as it should right?

All pulse shaping works ok .

I could have fake 4047's ? highly unlikely
what values did you use for the 4047 rc?

Its the same on both boards which almost rules out operator problem

Dave ...did you have any issues ?

I used 0.001uF and100k pots, that will give you 1kHz - 100kHz/ 2 = 500Hz to 50kHz on your Q output.

But you can use many combinations.

C is your control frequency, it has to be duty corrected before triggering the "A" and "B" oscillators. Remember "B" is reverse, if you have "B" ON at the same time as the others it is maybe the 4011.

The duty cycles I have mentioned are REF: only, they are a point to start eg. 25%, 16.66%, and 8.8%, it is the ratios that must be the same.

@ ALL

The TPU is a parametric amplifier, remember the words of SM "it naturally runs with gain", that is not where the energy comes from because all amplifiers have gain, but are not OU. It is the means of extracting the ambient energy through the plasma, all along the solenoid coils with reference to the core there is an infinite capacitance change within the plasma. When it gets to "B", C2 is switched into the circuit, and from there on the parametric pumping continues within the continuing cycles.

There is no race here as to whom first replicates, but everyone must realise how this works, it is an analysis of the circuit, and the most complicated part is realising the coil core varying capacitance. You can see it in the scope shots I have posted.

Here are some others

Regards

Mike

[Centraflow]

Perhaps a cro picture ? I have a ,c, and b  on together most of the time
Astable c controls all outputs frequency and phase  in sync.
Im not able to get b inhibited by c as it should right?

All pulse shaping works ok .

I could have fake 4047's ? highly unlikely
what values did you use for the 4047 rc?

Its the same on both boards which almost rules out operator problem

Dave ...did you have any issues ?

I think I see your problem, which is my problem.

As you are using the 4011 logic, the "B" 4047 should be wired gated astable as is "A" and not as complimented gating as I have drawn.

I have not tried this atm, mine is not using the 4011 logic, but it is an easy mode, let me know as I will have to change the schematic with the 4011, I just don't have time atm.

Regards

Mike

[Centraflow]

I have altered the schematic, attached below, this should resolve the problem.

Pissing down with rain so no outside work, well it's getting dark now anyway

Regards

Mike

[Centraflow]

I have been building a new unit and I have been documenting it as I have gone along, here is the first part.

Regards

Mike

[web000x]

Perhaps a cro picture ? I have a ,c, and b  on together most of the time
Astable c controls all outputs frequency and phase  in sync.
Im not able to get b inhibited by c as it should right?

All pulse shaping works ok .

I could have fake 4047's ? highly unlikely
what values did you use for the 4047 rc?

Its the same on both boards which almost rules out operator problem

Dave ...did you have any issues ?

Let me know if Mike's revision doesn't solve your issue.  I will make a video and go point by point with a scope for you so show you my signal propagation/logic levels.

I have not really had many issues with getting the oscillator circuit running.  The biggest issue I had was getting the frequencies in sync.  It is a little difficult if you are running the sequencer while trying to dial in the frequencies.  I found that moving the astable inputs of A and B to V+ while tuning 4047 frequencies is much easier since your frequencies are free to run.  Once you get your harmonic relations right, move the Astable inputs back to the output of the 4011 and just a few little pot tweaks will produce the waveform I posted earlier in this thread.

Dave

[3D Magnetics]

Thank you Gentlemen ..It was the gating wiring.

I was heading there ..I know and appreciate the issues with new circuits etc
the latest top view  circuit from mike does work

Back to the Real work of getting the adjustments right ..

I think that having a set up sequence for the free running 4047 freqs and pulse width's/order will be helpful for others .

[web000x]

If  anyone  can come up with a strategy that has the dds as reference source it will be better for learning than twiddling pots .

I will have one more go with some cowboy logic as see if i cant ruin the timing completely!

Have you familiarized yourself with the arbitrary waveform generator for the FY8300?  I am curious how accurately wave forms can be created.  If I were to attempt to create an arbitrary waveform that mimics the shape of B output, I could lock it in with A and C outputs (DDS full control, no 4047's) and adjust via phase and frequency on the dds and get fairly decent control.  The only real limitation would be adjusting the pulse widths of B output while remaining locked in with A and C frequency relations.  If wave forms are easily and accurately created via software, maybe a few different wave forms that mimic B could be created with varying duty cycles.  I've yet to install the software and have a look due to not having a decent PC at the moment.  I will load it on an older computer tonight and see what I find out.

I'm sure Mike's circuit works just fine.  It gets a little annoying to have to retune with each adjustment of the pots on the 40106 or 4047 oscillator circuit.  I know you can lock in adjustable relations on the DDS to get all of the outputs to move at once. 

Just thinking out loud.

Dave

[3D Magnetics]

there might be a limitation with the arbitrary function ..but with these things its best to try it and see..

Its possible that the right sync might be available by default ...lots to test there. the 3 channels do come from the same clock after all.

Im confident that a dds solution will be the future
there are some guys at eevblog writing code for the earlier units as the manufacturers released some machine control codes...might be worth a look there too.

Any way its good to be back on track with high confidence.

[web000x]

there might be a limitation with the arbitrary function ..but with these things its best to try it and see..

Its possible that the right sync might be available by default ...lots to test there. the 3 channels do come from the same clock after all.

Im confident that a dds solution will be the future
there are some guys at eevblog writing code for the earlier units as the manufacturers released some machine control codes...might be worth a look there too.

Any way its good to be back on track with high confidence.

I saw a video of the software for these DDS's creating an arbitrary wave.  You can input x,y coordinates so it would be doable.  I might give it a try.  It would at least allow for a frequency sweep of uniformity between the three outputs, duty would be locked unless I saved waveform triplets of varying duty cycles to toggle through.  I like being able to frequency sweep easily....

Glad you're back on track.  Measuring my toroid per Mike's recent pdf to compare and make similar calculations.  Trying to get to the fruit.

I recently split up with a crazy woman that didn't care too much for this research.  Glad to be back.

Dave

[3D Magnetics]

were you guys able to drive the 3210 from the 40106?

mine seems to have a current supply problem ..the 3210 led need more current .

a buffer stage needed?

[ainachara]

were you guys able to drive the 3210 from the 40106?

mine seems to have a current supply problem ..the 3210 led need more current .

a buffer stage needed?

no, there are no problems. For the LED in the 3120, approx. 10mA current should flow. The supply voltage of the 40106 then results in a corresponding series resistor. For example, I used 820 Ohm for 12V. Now at 3V from the ESP32 https://www.espressif.com/en/products/socs/esp32 as a signal generator 270 Ohm.

[web000x]

Lindsay,

Do you have the software for the FY8300?  I cannot find my disc anywhere.  I have the JDS6600 software for my other unit but can't locate the FY8300..

Dave

[3D Magnetics]

Here it is...
You need to download the user guide and its all there ...not exactly clear, i had the same issue

Th serial protocol is there too ...im not a great coder but it looks very promising ....if it does what it seem to indicate.
heck, I cant even wire up a led properly on some days!

http://en.feelelec.com/index.php?case=archive&act=list&catid=6


Edit
Another point that might be useful is to fit a ground lift switch which switches the main bnc earths from mains ground to floating ...with all the cautions that go with it .
some use a high value resistor.
When it is hooked up to the computer via usb the ground is through the computer...unless it is a floating laptop.
The earth loop may otherwise pick up stray interference, which we generate .

[3D Magnetics]

no, there are no problems. For the LED in the 3120, approx. 10mA current should flow. The supply voltage of the 40106 then results in a corresponding series resistor. For example, I used 820 Ohm for 12V. Now at 3V from the ESP32 https://www.espressif.com/en/products/socs/esp32 as a signal generator 270 Ohm.

Yes.  My brain is worn out ,i must have a breadboard problem with the led wiring ...too many different supplies and grounds on one pcb..

Here is a little tip

The 40106 does not like 15 volts when driving leds  1.2k resistor ..fried them when tuning .
all logic and generators now at separate 5v and stable now .

[Centraflow]

Now who can tell what they see in this scope shot

Regards

Mike

[Centraflow]

When looking at resonance what are you looking at? yes, the classic ring DOWN of the resonant frequency.

The question is can there be a ring UP? if so where is the energy coming from?

Regards

Mike

[web000x]

Now who can tell what they see in this scope shot

Regards

Mike

Ummm....  Positive energy coefficient.   

[web000x]

Hi Toni

No, attached is revision 2.9

As the input from A and C charge chokes is a pulse and no return is required, unlike a half-bridge, the diodes let through this pulse but do not allow, in the case of "C" mosfet, a short of the "A" pulse through that mosfet if the timing is not right.

Regards

Mike

Mike,

Are you using this circuit?  It seems C2 ceases to be resonant when the A & C drain diodes are present. I was trying to figure out why no current was flowing through my B transistor and then I got bit by the cap!    It would seem difficult to find resonance before connecting a load..

Dave

[ainachara]

Now who can tell what they see in this scope shot

Regards

Mike

Ok, Mike, what I see in this picture is:
1. a period of approx. 1.4 s, i.e. a frequency 0.74 Hz needle pulses of approx. 350 mVpp
2. in between 5 resonance points at a distance of 260 ms, i.e. a frequency of 3.846 Hz
3. and, this is interesting, the amplitudes of the 5 resonance points increase.
Now the question for you: with which circuit and at which point did you measure this?
And how do you use this principle on a larger scale so that it is useful?

Regards
Toni

[Centraflow]

Toni

I'm away from home in Valencia to collect my wife's grand daughter, will be back this afternoon. I was testing the new carefully built Toroid for posting pictures etc.

What nobody looks at is "sub harmonic resonance", and without taking a second look at resonance scope shots you can easily pass by a shot like that.

The frequency is close to the 2nd sub harmonic of 7.83.

Regards

Mike

[Centraflow]

@ Toni

OK the signal generator is set to square pulse at 1kHz 10v amplitude into a 10Mohm resistor.

The channel 1 probe is connected to the TPU side of the resistor.

The resistor is connected to the "a" coil where normally the drain of the "A" Mosfet is connected.

The loop (core) ground is connected to the "b" coil where normally the drain of the "B" Mosfet is connected, the scope ground is also connected here along with the signal generator ground.

I have redone the test now and all is the same, scope settings are in the photo posted before..

Below is another scope shot with the same connections but different scope settings.

I have also taken a 5sec video of this which shows the CCW movement. My TPU is connected that way looking from above the toroid.

Hope that helps

Regards

Mike