Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated

[seaad]

Just interested of his background and from where he got his skills
Arne

[Jeg]

There are two factors which make sense wiring transistors and diodes in paralel: lowering resistance and eliminating generated heat.
Instead of wasting energy as heat I am also doing paralel diode connections in experiments when frequency and internal capacitance do not interfere.

Cheers!

Hi Arunas
I don't think this is the case here. Pierre uses diodes of a high breakdown voltage  and of enough current max rating for this job. My opinion is that Pierre made just a simple mistake here presenting both diodes of the same direction while it is of the opposite.

Regards

[d3x0r]

reading in the past 'magic numbers'
the pairing is more like in this image as opposed to opposed...
there was also an early magnet simulation by pmgr would like to see that with a wider pitch on the poles... (n/s of fields separated by 60 degrees Nevermind found here, and here) .  There was a later supposed animated gif, but I couldn't get it to animate even in VLC.

And Re the diodes... 72 diodes. 72 relays, 1 feed from diode board to + .  No - (negative is on the other side of the coil/relay)   - -> relay -> (fork to diode) -> coil -> positive.  (electron current from - to +, when the relay opens, it wants to keep pulling - from the diode/relay side, which will pull eventually against the positive voltage).

But I don't really know how actually 2 relays + 1 coil = reversable coil... this scheme works, if either one or the other relay is activated, power causes a magnetic current.  But, when both are open, there is just a free-wheeling current in the coil and the diode does nothing...  Other ways I tried to hook it up, the diode ended up in parallel with the coil, which again ends up in a lossy short-circuit loop in the coil...

So I suppose each coil is charged...
I haven't finished going through the middle pages (20-40 and 1-20)


---
I saw that 'pairing' diagram a few time... and while it's true that 1 and 6 are pared, and then 6 and 11 are paired, this does not mean that 1 and 11 are paired.


I colored paired coils (in a span, it's incorrect, span should have been 5, but this drew better) ... so when 2 is on because it's paired with 33, it's the opposite polarity from when 2 is on and 7 is it's pair.....
I highlighted under each with blue and red indicating the poles of the pairing.  The blue is always to the left of the red.
Doing the stepping better would have shown the rotation better....


(other 'fixed') but should be parallel not series...


----
Image with pitch of 5 marked.  (for 36 coils).  pitch = ( ( coils / 6 ) - 1 )  ; although you can re-sequence it and use just 2 poles...
(too bad it was so late after  Reply #395 on: March 21, 2018, 05:43:39 PM »    "In Pierre's 36 slot he obviously has 6 coils or slots per pole and you can clearly see in the picture below his center core is exactly 6 slot wide."  ... yes, but that's because 2 coils are on at a time so it's (coil-slots + 1) = 6 (the width of the core)

[listener192]

Hi Arunas
I don't think this is the case here. Pierre uses diodes of a high breakdown voltage  and of enough current max rating for this job. My opinion is that Pierre made just a simple mistake here presenting both diodes of the same direction while it is of the opposite.

Regards

No, there is only a positive return to the super caps.
There is no negative wire going to the diode board.
The current path for recovery is through the low side switch then through the chain of coils and through the diodes to the + rail.

Regards

L192

[Jeg]

No, there is only a positive return to the super caps.

Yes you are right. Hadn't notice that.