1932 Coutier Device

I was 2-3 weeks busy and could not handle the project in the meantime. I also need to build a stronger winding device for thicker wire.
Ok, I studied the scheme a little, the mistake at 10:30 on the left is clear (on the right is ok).
Is there any information regarding the number of primary / secondary turns or the amplification factor, voltages, etc.?
Yes, at first glance it seems that the primary coils (15 pcs) are connected in series and the secondary ones in parallel (another 15 pcs).
It's strange that all the coils have the same section, a kind of toroid-shaped core.
More to be studied ...
All the best.
EmilP

Very nice drawings indeed. Nice that you have this in a bigger size, because i downloaded this, but it was unreadable.

Would love to see the big one .

kind regards.

Peter

WOW... Thank you MM. Very cool..

This video is NOT a Bi-toroid Transformer. It is the Coutier 1932 with a primary
and two secondaries. I had the same material as in the video at home. I built it
and measure it. It had four layers of copper wire as the primary, but 5 layers
of copper wire on the secondaries. I thought by using more turns on the secondaries
I could rectify the outputs and lose 1V2 over the rectifying diodes and loop it to
a 5-20 khz oscillator. The cores in the video are ferritecores from an old TV-Set.

When measuring the output I was very dissapointed to find the voltage loaded on the
secondaries was only half of the input voltage and I measured both the input and
output voltage*ampere and found the coutier transformer was less then 100% effective.

Try to build the coutier transformer inverted. Double the size of the secondaries
core compared to primary core. I have asked for a spanish to english translator at
groups.io/g/EVGRAY you are invited to join. Write: "I just joined" in the
message field.

Best Wishes, Hermes

So back on track, how is it going Emil have you made any forward progress.??

the cores are pretty much straight forward as we should all know the surrounding cores cross section combined area should match that of the center core with a little headroom. we also know that each coil set has the same winding count as it's center core in each group as a one to one ratio which are paralleled into the next coil/core group. so then it is logical thicker wire is used as current ramps up and less amper turns per core /coil group.

so this leaves us with the feedback isolation portion of the circuit and the actual oscillation or excitement of the original coil/core. if one so chooses to use mechanical oscillation as in the original patent then the use of a Mallory oscillator type setup is deployed. if electronic deployment is used then one could use a pure sine wave inverter or something similar to get the excitation and the use of an isolation transformer. my suggestion would be to use a toroid transformer for the isolation as toroid's are much more efficient then say a microwave transformer with square corners.

of course there will be some trial and error involved but that is the least expensive part as the acquiring of the cores and excitation/isolation are the biggest investment.

regards,
Marathonman

Yes that looks like a good idea with that counter, very low power indeed. it will be difficult cutting Mot's as a whole but it can be done. i think the whole key is to just take your time cutting. the bast way i know to cut them is with a metal band saw. it can be cut this way without clamps as the cutting force is always down but supported underneath. the next would be to clamp it or in vice and clamp then cut between those two. or a more time consuming step is to completely disassemble them and cut with sheers then epoxy back together like i do with my cores.

to take them apart to start the weld below has to be cut, granted some material is lost but somethings are unavoidable.



once the weld is cut the core can be taken apart. as i said you can cut at the lines shown or take it completely apart. just take your time as it is quite easy to flare or bend the laminations. cutting across the width of the lamination is best as cutting across the depth will flare or bend them.



another thing is pay attention to the grain of the metal as the above lamination has the grain going side to side and the end piece is up and down. there is no real easy way about this so all i can say is dive in the best way you can think of but just do not get in a hurry. even if you do slightly bend the laminations you can always epoxy then put it in a clamp or vice over night.

when estimating your cores you can weigh the bare core then estimate the lbs per va rating.  as an example say i had an 1,100 watt core and it weighed 10 lbs then it would have a va rating of 110 per lb of lamination which will help with the calculations of the output of each core. each core should weigh the same or as close as possible.

Mot's are a pain in the backside but it is completely doable for sure. remember even if the Mots are of a different rating or manufacture they generally use the same material for all mots so the concentration of the weight is more important then the material it's self.
remember folks Mot's can be used for the Figuera device also and the 1882 device i have posted.

Regards,
Marathonman

First of all, I wish you a "Happy New Year!" and I wish you that 2020 will be a better year, full of joy, success and health!

The observations you made above are very correct. The loss of magnetic flux between the central core and the satellites (20-30%) must also be considered.
When the satellites are short-circuited for a few seconds, the core must not saturate. If the core becomes saturated in secondary the current is zero. So I can test if there is enough iron core.

All the best.
EmilP