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

[TinselKoala]

3 days ago I sent a message to Pierre asking him to confirm if his coils are powered in series. I've just received a reply from him which I translated and below is his original French message.

Fr. Il y a 3 jours j'ai envoyé un message à Pierre lui demandant de confirmer si ses bobines sont alimentées en série. Je viens de recevoir une réponse de lui que j'ai traduite et ci-dessous est son message original en français.

in my assembly there is 6 coils in series except that there is one that I feed for a fraction of a second so not to lose the magnetic field, so it leaves me with 5 coils operating in reality, example: 1 and 2 are powered by the positive and then 1 goes off to allow the magnetic flux to move etc. So in your case (Pierre is referring to my 30 slot) there should be 4 coils operating using the program I gave you which is good. The only thing I may of modified is the value x and y and played a bit with the speed. Everybody is making too much of a deal with the frequency. I already mentioned that the frequency meter may not of been correct. We can see all the uneven switching noise on the oscilloscope which affected the frequency meter. I adjust the potentiometer but most likely the frequency meter did not read the right frequency. Instead of having a discussion on frequency, people should be working on rotating the magnetic field. Then they can start playing with frequency and increase it to the desired speed.

Fr. dans mon montage il y a 6 bobine en série sauf qu'il y a une que j'alimente pendant une fraction de seconde pour ne pas perdre le champ magnétique donc il me reste 5 bobine en fonction en réalité ex:1 et 2 sont alimenté par le positif et  puis 1 s'éteint pour déplacé le flux magnétique ainsi de suite dans votre cas il devrait  rester 4 bobine en fonction  en passant le programme que je vous ai donner est bon la seul chose que j'ai peut être modifier est la valeur x et y j'ai jouer un peut avec la vitesse et puis le monde s'en font vraiment trop pour la fréquence ,j'ai déjà mentionner que le fréquence de mon meter n'a peut-être pas  la bonne valeur avec  ce que j'ai vue a l'ociloscope  a cause des parasite la fréquence était anarchique j'ai ajuster le potentiometre mais il se peut que le fréquence meter n'a pas lue la bonne fréquence  avant de  parler de fréquence il devrait se concentrer par faire tourné le champ magnétique après ils pourront s'amuser avec la fréquence il suffit juste d'augmenter a la vitesse désirer

Has there been another Arduino program received from Pierre? Because the one I saw and analyzed does not appear to do this.

[listener191]

Has there been another Arduino program received from Pierre? Because the one I saw and analyzed does not appear to do this.

Hi TinselKoala,

These lines in Pierre's original code, as an example for the first and second coils support an overlap.

digitalWrite(1,HIGH), digitalWrite(13,HIGH), digitalWrite(25,HIGH);   // turn the LED on (HIGH is the voltage level)
  delay(x);              // wait for a second
  digitalWrite(2,HIGH), digitalWrite(14,HIGH), digitalWrite(26,HIGH); // turn the LED on (HIGH is the voltage level)
  delay(x);              // wait for a second
  digitalWrite(1,LOW),  digitalWrite(13,LOW),  digitalWrite(25,LOW);     // turn the LED off by making the voltage LOW

Regards

L192

[listener191]

Extensive running of L298N half bridges (parallel transistors) at 4A into one set of 6 coils, resulted in a device failure.

The running temperature was only 40 deg C for the high side switch.

4A into two sets of 5 coils via one source and one sink half bridge, for my 30 slot stator, would likely be a minimum for me, so I think the L298N will prove not fit for purpose.

I want to evaluate the BTN7960 board carefully, as the relatively low rail voltage limit of 28V may be limiting if the clock rate is raised to say a 1ms period, as higher voltage will be required to get the current rise within the period.


Regards

L192

[Jeg]

Why were Pierre's diodes only connected to the plus rail? FR/   Pourquoi les diodes de Pierre étaient-elles uniquement connectées au rail plus?

Because he made a mistake when drawing. That simple. One diode looks to the positive rail and the other to the negative. Like all clamps.

What i don't quite understand is why there are about 100 steps between two peaks (whole period). Look at the drawing. There should be just 12 switching events until the full period to be completed. Any idea?

[TinselKoala]

Hi TinselKoala,

These lines in Pierre's original code, as an example for the first and second coils support an overlap.

digitalWrite(1,HIGH), digitalWrite(13,HIGH), digitalWrite(25,HIGH);   // turn the LED on (HIGH is the voltage level)
  delay(x);              // wait for a second
  digitalWrite(2,HIGH), digitalWrite(14,HIGH), digitalWrite(26,HIGH); // turn the LED on (HIGH is the voltage level)
  delay(x);              // wait for a second
  digitalWrite(1,LOW),  digitalWrite(13,LOW),  digitalWrite(25,LOW);     // turn the LED off by making the voltage LOW

Regards

L192

Yes, I realize that. After all, I have the sketch running on a Mega right in front of me.
What I am questioning is the number of coils that are being turned on according to the quote from Perre. Does this posted sketch turn on 4 coils at a time?

Another issue that has never been cleared up, as far as I can tell, is the timing irregularity present in that first sketch posted from Pierre.  Is this intentional, or should there be an even, smooth flow with equal timings all around the ring? If it is intentional, what is the reason and does the demonstrated OU effect depend on it? If it is not intentional... then wtf?