This is it !

[Farmhand]

I don't have any PM's. I am still sticking with the program of ours. Nothing new has been said. Just like you said Greg. We are already aware of all this.

I am patiently waiting for the video. I have not nor will I have any secrets about this between us. THats not what I meant in my other post, merely that if he had info to share but didn't want to share in public he could PM any one of us, we are a team ! Aren't we ? Haha looks like we are like the three Musketeers or the three Amigo's but I won't say the three stooges'.

I also just recently stated that we don't seem to have left any stones unturned. I also mentioned The EPG before too. It would be benificial if people did read the whole thread but we can't controll that.

Anyway nothing has changed for me since yesterday, and I have a funny feeling that there will not be any great revelations about this, I think we have it pretty well covered.

I feel very lucky to be able to try your circuit Greg and I can't wait I think it will work just fine. You do very good work.

Cheers

[gmeast]

no ... "Stooges" is fitting

I don't have any PM's. I am still sticking with the program of ours. Nothing new has been said. Just like you said Greg. We are already aware of all this.

I am patiently waiting for the video. I have not nor will I have any secrets about this between us. THats not what I meant in my other post, merely that if he had info to share but didn't want to share in public he could PM any one of us, we are a team ! Aren't we ? Haha looks like we are like the three Musketeers or the three Amigo's but I won't say the three stooges'.

I also just recently stated that we don't seem to have left any stones unturned. I also mentioned The EPG before too. It would be benificial if people did read the whole thread but we can't controll that.

Anyway nothing has changed for me since yesterday, and I have a funny feeling that there will not be any grat revelations about this, I think we have it pretty well covered.

I feel very lucky to be able to try your circuit Greg and I can't wait I think it will work just fine. You do very good work.

Cheers

Choke ....we will be upon needing specifics about the choke. In video0, W shows the trace and then says " ... and we use the choke to filter (that) ...", etc. Crazy mysterious.

I just like to rant at these types. He may have a 'working knowledge', but me no thinks so, but not much to offer, see?

With in the next day or so, I will hook up at least one driver IC and fiddle with it. I still don't know how it's going to work hooked to 12VDC, and the MOSFET needing an inverted pulse from +12VDC to GND when my input to the driver IC comes from only 5VDC ... what do I send an inverted signal to the driver from 4VDC to GND? I can already see the purple smoke pouring from the chip.

Oh well, I'll figure it out ... someone please explain. Thanks in advance.


Public is at:
YouTube - Gregs_Multi-Phase_ Controller0_pub.MPG
Private is uploading

Later

[Farmhand]

Choke ....we will be upon needing specifics about the choke. In video0, W shows the trace and then says " ... and we use the choke to filter (that) ...", etc. Crazy mysterious.

I just like to rant at these types. He may have a 'working knowledge', but me no thinks so, but not much to offer, see?

With in the next day or so, I will hook up at least one driver IC and fiddle with it. I still don't know how it's going to work hooked to 12VDC, and the MOSFET needing an inverted pulse from +12VDC to GND when my input to the driver IC comes from only 5VDC ... what do I send an inverted signal to the driver from 4VDC to GND? I can already see the purple smoke pouring from the chip.

Oh well, I'll figure it out ... someone please explain. Thanks in advance.


Public is at:
YouTube - Gregs_Multi-Phase_ Controller0_pub.MPG
Private is uploading

Later

There is a rail to rail pulse from the output of the driver, same as or similar to the simple driver setup I tried. I think that is what you mean. The driver chip can pass up to 9 amps from the12v + rail bounce it off the Mosfet gate and then ground it. It can be an amplifier if used like one. I think.If you fry anything it will maybe be the mosfet switch.

The mosfet can use positive or negative pulses I think too, not sure about that.

7imix should clear that up.

Cheers

[Farmhand]

Are the phases trying to lock right after one another ? Because I think that will be as good as in unison probably better at some frequency. If we want that.

I was thinking that as the frequency gets higher we may want to pull the phases closer together in time. There may even be a point at which going the other way spacing them into the other phase area will be benificial. It's hard to explain i'm seeing it in moving pictures. That will take experimenting to see the best results from whichever combinations. Which will be a piece of cake with that setup. There should be some phase pattern or frequency combo that can make the next pulse just clip the back end of one of the existing intensities to slingshot it maybe, pull it or push it shouldn't matter.. Will be fun and interesting to find out.

I feel a bit like a third wheel at the moment . Hahaha

Cheers

[7imix]

Sorry can't clear that up!

[Farmhand]

OK i'll confirm it then

Metal Oxide Semiconductor MOSFET Tutorial

We saw in the previous tutorial, that the gate of a JFET must be biased in such a way as to forward-bias the PN-junction but with a insulated gate MOSFET device no such limitations apply so it is possible to bias the gate of a MOSFET in either polarity, +ve or -ve. This makes MOSFETs especially valuable as electronic switches or to make logic gates because with no bias they are normally non-conducting and this high gate input resistance means that very little or no control current is needed as MOSFETs are voltage controlled devices. Both the P-channel and the N-channel MOSFETs are available in two basic forms, the Enhancement type and the Depletion type.

I think that is what the bolded text means.

[gmeast]

set-up

Are the phases trying to lock right after one another ? Because I think that will be as good as in unison probably better at some frequency. If we want that.

I was thinking that as the frequency gets higher we may want to pull the phases closer together in time. There may even be a point at which going the other way spacing them into the other phase area will be benificial. It's hard to explain i'm seeing it in moving pictures. That will take experimenting to see the best results from whichever combinations. Which will be a piece of cake with that setup. There should be some phase pattern or frequency combo that can make the next pulse just clip the back end of one of the existing intensities to slingshot it maybe, pull it or push it shouldn't matter.. Will be fun and interesting to find out.

I feel a bit like a third wheel at the moment . Hahaha

Cheers

Hi Farmhand,

This circuit is so basic and uses the fewest parts possible to accomplish a difficult task. The trade off for this is that nothing is proportional. Everything is quite independent of everythingelse. If you change the clock frequency, you are changing the pulse width. You adjust the 'operational frequency' by changing the counter recycle with the dip switches. Now you have to readjust the phase angles ... again, dip switches, but it's not that hard since the dip switch inputs are a binary number and everything is just arithmetic. I told you earlier, I intend to build a graph or chart that will let you look at the axiis and go:

"I want this, so set the binary on the clock to this ... now go over to this and set the binary on the frequency (recycle) to this, then go up this side and select # of phases and then select phase angle, set those dip switches ... etc" and tah dah! it's set. Maybe an online calculator or other simple 'software calculator'.

Once calibrated, just set the dip switches for your pulse width, frequency, # of phases and phase angles ... it's done. That's been my plan for this thing all along ... I just didn't know 'exactly how to do it' but I do now. You won't even need a scope ... but everyone here EVEN puts a scope probe on a baked potato ... come on ... you know it's true ... we all probe everything. All those Space Aliens got nothin' on us!

Yes, you should be able to sneak up on another transient and clip its ass.

You two, check you PM's

Later

[Ordo_Ab_Chao]

....errr

sorry...will not comment on YOUR thread

[gmeast]

driver chip

Hi everyone,

I hooked up the driver chip and ran it. The vid of that is here:
YouTube - Gregs_Multi-Phase_Controller2_pub.MPG

Enjoy.

Later

[gmeast]

overshoot/undershoot

Hi everyone,

I hooked up the driver chip and ran it. The vid of that is here:
YouTube - Gregs_Multi-Phase_Controller2_pub.MPG

Enjoy.

Later

In the video you can see overshoot, undershoot and 'ringing' both above and below the rails. I have fixed that with some strategically placed capacitors per the data sheet and application notes for the UCC27321 gate driver IC.

A pic of that is here:
http://02d1852.netsolhost.com/radian...iver_chip0.png

Later

[7imix]

Awesome gmeast. I also hooked up the driver chip and ran it, but I just hooked it up straight to the clock and ran it at a lower speed. It performed beautifully, but I was missing the filter caps and maybe some other protection mechanism and when I hooked it up to a small toroid, the huge voltage spikes fried the driver. It died for a good cause though, I measured the whole circuit drawing 5 milliamps at 7.5 volts and the spikes in the coil were huge, up to almost 100 volts. This circuit is incredibly efficient and productive, now I need to figure out how to protect the driver properly.

[gmeast]

Awesome gmeast. I also hooked up the driver chip and ran it, but I just hooked it up straight to the clock and ran it at a lower speed. It performed beautifully, but I was missing the filter caps and maybe some other protection mechanism and when I hooked it up to a small toroid, the huge voltage spikes fried the driver. It died for a good cause though, I measured the whole circuit drawing 5 milliamps at 7.5 volts and the spikes in the coil were huge, up to almost 100 volts. This circuit is incredibly efficient and productive, now I need to figure out how to protect the driver properly.

That's great. What driver are you using? I bought the UCC27321's ... the Inverting ones and I should have bought the NON-Inverting drivers. Don't know what I was thinking. I had to go back to my old rig and run it with the IRF540Z's. Low-side, N-Channel MOSFET = positive-going signal. Also, you have to use a resistor between the driver output and gate, if you don't then bye bye driver ... lost 2 of them because I didn't size it right. It's in the application notes ... I think.

[Farmhand]

Good job Gmeast Thats a big difference. Couple of small caps. I think you have a cap on the power input pins, I think I seen it.


7imix, 5 Ma 7.5 volts nice and it made spikes that fried the driver. Excelent. Spikes caught you off guard. Destructive little devils.

You really should load the recovery when you pulse a coil in a way that makes spikes, even if it is with a 12 v light bulb or something, I always use a battery (as long as I don't forget) haha. It's just like how an SSG fries it's transistor when the battery is disconnected. Gotta give that spike somewhere to go or it can go back through the circuit. Kaput.

A neon connected one leg to the negative of the coil and the other to ground should ground anything over 80 volts or so. I've measured spikes 800 volts with my SS Bedini while the charging battery is disconnected, they will fry stuff if they can't escape the circuit. I think I have fried a few different chips like that forgetting to connect the recovery. And a lot of transistors with my bedini stuff. It started to get tiresome replacing them. One problem for SSG's is there is no cap in the circuit, if there is some caps on the circuit it can help protect from the spikes. Better to remove it though.

It performed beautifully, but I was missing the filter caps and maybe some other protection mechanism and when I hooked it up to a small toroid, the huge voltage spikes fried the driver. It died for a good cause though, I measured the whole circuit drawing 5 milliamps at 7.5 volts and the spikes in the coil were huge, up to almost 100 volts. This circuit is incredibly efficient and productive, now I need to figure out how to protect the driver properly.

[Farmhand]

That's great. What driver are you using? I bought the UCC27321's ... the Inverting ones and I should have bought the NON-Inverting drivers. Don't know what I was thinking. I had to go back to my old rig and run it with the IRF540Z's. Low-side, N-Channel MOSFET = positive-going signal. Also, you have to use a resistor between the driver output and gate, if you don't then bye bye driver ... lost 2 of them because I didn't size it right. It's in the application notes ... I think.

Yes that is true for the gate charge to get to ground it will need to go back through the driver chip, unless a resister or PNP transistor is used externally to the driver chip. In order to use a PNP transistor a 1n4148 diode must be used so the PNP base can go to the diode-annode and the PNP collector to the diode cathode the PNP emmiter to ground, it won't work without the diode. If a resistor is used it would influence the pulse width, wouldn't it ?

Cheers

[7imix]

That's great. What driver are you using? I bought the UCC27321's ... the Inverting ones and I should have bought the NON-Inverting drivers. Don't know what I was thinking. I had to go back to my old rig and run it with the IRF540Z's. Low-side, N-Channel MOSFET = positive-going signal. Also, you have to use a resistor between the driver output and gate, if you don't then bye bye driver ... lost 2 of them because I didn't size it right. It's in the application notes ... I think.

Yeah it must have been the missing resistor, thanks. I'll read the app notes.

I got the 27321 in inverting, non inverting, soic, and dip... I was using one of the soics, not sure which.