Author Topic: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated (Read 433033 times)

pmgr · « **Reply #165 on:** March 31, 2018, 03:16:41 AM »

Luc, one more question I have about the last picture you posted is how Pierre controls the direction of the current. The Arduino appears to only control the transistors that control the relays, but what determines if the top relay turns on or the bottom relay? I would assume one transistor controls both at the same time.

E.g. in your picture, let's say the left two relays with the forward flowing arrows are turned on with transistor 1 (so coil 1 on), then if for that particular coil he wants to switch current direction, what transistor(s) would he need to turn on? If he turns on the second transistor, it would give the same forward flowing current in the second coil, not reverse current. He would need to turn on the GND connection of coil 36 (controlled by transistor 36).

If he turns on the first transistor (so he gets the GND connection for second coil for negative flow), he would also need to turn on the third coil. Then he will get reverse current in second coil, but the third coil will also be on in forward direction. And he would have transistor 1 and 3 on at the same time. His Arduino code doesn't show any evidence of that kind of transistor switching.

PmgR

e2matrix · « **Reply #166 on:** March 31, 2018, 09:14:36 AM »

I'm not sure if these were mentioned before but in cruising eBay I notice these 8 channel 8 relay module boards for Arduino for only $4.96 (from Hong Kong) or same one from a U.S. seller for $7.95 which appear to be identical to the 8 relay boards in Pierre's video as seen at 6:16 into his second video which showed it completely self running. You can easily find them on eBay by these item numbers: eBay item number:172396554144 for the $4.96 one from H.K. and eBay item number:231682185473 for the $7.95 one from the U.S. I'm quite sure they are identical to the one's in Pierre's video for anyone wanting to do a relay type replication. Nine of the boards would be needed for your 72 relays which only comes to about $45 total with shipping - not bad for all that many relays with screw connections and rated 10 amp at 250 VAC and Arduino ready.

r2fpl · « **Reply #167 on:** March 31, 2018, 09:23:32 AM »

Building H-Bridge, I suggests used 2 sets:

AOD609: N/P-MOSFET; http://aosmd.com/pdfs/datasheet/AOD609.pdf
or
FDD8424H: N/P-MOSFET; http://www.mouser.com/ds/2/149/FDD8424H-92711.pdf

e2matrix · « **Reply #168 on:** March 31, 2018, 09:24:19 AM »

Quote from: keykhin on March 31, 2018, 01:52:25 AM

Hi everyone! This is my first post here. I'm following this thread for a while and I want to give you some advice. That L298 is suitable for RC's and small stepper motors used in 3D printers or small CNC machines. In this DZ particular case you have to deal with coils that have low resistance and high inductance. Use specially designed mosfets with logic level drive and very low on-resistance like IRLR2905 or IGBT's like IXDH35N60B from IXYS Corporation or equivalent. But I advice everyone for the first build to use relays. As a personal project I have in my mind to design a simple and low cost sequencer based on a PIC microcontroller. Using one Arduino for this simple job is like killing a mosquito with a sledgehammer. I will return when I finish the code. Cheers, K.[/size]

It seems Arduino's have gotten so cheap ( $10 to around $18 for a Mega 2560 R3) that they are as cheap or even less than many PIC controllers and can do a lot more with them if you decide to use it for something else. But if you are more familiar with PIC controllers and can get the same results thats all good too.

listener191 · « **Reply #169 on:** March 31, 2018, 09:25:18 AM »

Quote from: gotoluc on March 30, 2018, 11:51:56 PM

That video of his complete circuit was him sharing a possible solution to eliminate the WWW wave he was seeing on his scope.

Pierre's instructions and diagram prior to making that video.

Eng. About the sequence of the arduino, you will have to turn on one coil and then the other and then turn off the first coil otherwise you will break the magnetic field. The goal is not to turn off the magnetic flux but only to keep it moving ex: 1 on, 2 on, 1 off, 3 on, 2 off etc.

Fr. pour de ce qui est de la séquence de l'arduino quand vous ferez la séquence vous devrez allumer une bobine et puis l'autre et par la suite éteindre la première sinon vous allez coupé le champ magnétique le but étant de ne pas éteindre le flux seulement le déplacer ex:1 on, 2 on, 1 off ,3 on, 2 off etc.

I don't know but I read that and see the diagram he provided with those instructions as coils are individually powered and not 5 or 6 powered in series in a row.
However, there are 6 poles, so at all times there are 6 coils powered plus another 6 for the overlap time.

Regards
Luc

Hi Gotoluc,

I believe that diagram Pierre made was just to demonstrate the switching principle, the larger diagram in the video showing all the coils, was made to show that 6 coils were energize in series.

Regards

L192

listener191 · « **Reply #170 on:** March 31, 2018, 09:40:12 AM »

Quote from: pmgr on March 31, 2018, 03:16:41 AM

Luc, one more question I have about the last picture you posted is how Pierre controls the direction of the current. The Arduino appears to only control the transistors that control the relays, but what determines if the top relay turns on or the bottom relay? I would assume one transistor controls both at the same time.

E.g. in your picture, let's say the left two relays with the forward flowing arrows are turned on with transistor 1 (so coil 1 on), then if for that particular coil he wants to switch current direction, what transistor(s) would he need to turn on? If he turns on the second transistor, it would give the same forward flowing current in the second coil, not reverse current. He would need to turn on the GND connection of coil 36 (controlled by transistor 36).

If he turns on the first transistor (so he gets the GND connection for second coil for negative flow), he would also need to turn on the third coil. Then he will get reverse current in second coil, but the third coil will also be on in forward direction. And he would have transistor 1 and 3 on at the same time. His Arduino code doesn't show any evidence of that kind of transistor switching.

PmgR

Hi Pmgr,

Yes you are correct, each high side relay coil is also linked to a respective low side relay coil, (6 coils further around the loop), so one line switches both hence only 36 control lines are need with his scheme.

I think this is what he started with and what he showed us however, as there appear to be less than 36 wires going to the stator, I wonder if he has paralleled the 3 north pole together and paralleled the 3 south poles together? That would reduce the wires to 12. The number of wires appears to be 28 but there are at least two wires going to the same contact on one of the connectors.

It would take some thought to realize what the effect of that would be.

Regards

L192

cheors · « **Reply #171 on:** March 31, 2018, 09:47:40 AM »

Pour faire tourner le champ magnétique régulièrement il faut activer 2 bobines adjacentes L1, puis L1 et L2, puis L2, L2 et L3,....
Or les bobines sont alimentés + - pour la première et - + pour la seconde.
Comme on veut le même champ pour le renforcer, il me semble que l'on doit bobiner L1 dans le sens horaire et L2 dans le sens anti-horaire.
Ainsi les bobines impaires seraient CW , les paires CCW.
A -t-on déjà parlé de çà ? Evident ou pas ? J'ai manqué quelque chose ? le secret de Pierre ?

To rotate the magnetic field regularly it is necessary to activate 2 adjacent coils L1, then L1 and L2, then L2, L2 and L3, ....
But the coils are powered + - for the first and - + for the second.
As we want the same field to strengthen it, it seems to me that we must wind L1 clockwise and L2 in the anti-clockwise direction.
Thus the odd coils would be CW, the even coils CCW .
Have we already talked about that? Obvious or not? I missed something? the secret of Peter?

r2fpl · « **Reply #172 on:** March 31, 2018, 10:55:04 AM »

They can be all CW or CCW does not matter. For convenience, there could be CW and CCW less connection distance. In the configuration, all CW can change the ends of every CW then there will be reverse polarity.

example:
all CW or CCW = A1 .... B1, A2 ... B2, ..... the connection will be: A1 + B2 and B1 + A2
for once CW and CCW = A1 .... B1, A2 ... B2, ..... the connection will be: A1 + A2 and B1 + B2

Of course, it depends on the subsequent connections between the coil orders.

T-1000 · « **Reply #173 on:** March 31, 2018, 12:36:22 PM »

In regards to sequence, each group have 6 coils N and 6 coils S for three phase rotation. Only 4 coils used at once before flipping over to other set.
So the the three phase sequence for moving magnetic field would be in coil positions:
1) V+2 V-5 and V-31 V+35
2) V+2,V+3 V-5,V-6 and V-31,V-32 V+35,V+36
3) V+3 V-6 and V-32 V+36
Then continuing to flip around ring over next shifted sets of coils.

pmgr · « **Reply #174 on:** March 31, 2018, 01:20:47 PM »

Quote from: listener191 on March 31, 2018, 09:40:12 AM

Hi Pmgr,

Yes you are correct, each high side relay coil is also linked to a respective low side relay coil, (6 coils further around the loop), so one line switches both hence only 36 control lines are need with his scheme.

I think this is what he started with and what he showed us however, as there appear to be less than 36 wires going to the stator, I wonder if he has paralleled the 3 north pole together and paralleled the 3 south poles together? That would reduce the wires to 12. The number of wires appears to be 28 but there are at least two wires going to the same contact on one of the connectors.

It would take some thought to realize what the effect of that would be.

Regards

L192

Someone else stated that there are actually 5 green connectors in the video (hard to see), not four. So that would make 4x8 + 4 = 36 connections to the stator. Also, his boards show labels going up to 36.
PmgR

seaad · « **Reply #175 on:** March 31, 2018, 01:39:24 PM »

5 conns
But still Pierres repetition frequency is 5 Hz !

r2fpl · « **Reply #176 on:** March 31, 2018, 02:07:34 PM »

I found problem L298n.

When connect 2 channel (ex. coil1 and coil2) Voltage is different beetwen them.

I change L298n to another new. I changed coil 1-2, 2-1. Still channel 1 is more voltage to 2 channel less voltage !

1 = 0.56 ohm
2 = 0.56 ohm

When driver no load, voltage is equal.

Solution:
join 2 channel together and make one L298n for 1 coil

calibrate voltage used PWM. this method takes 2 pin arduino more. Total 6 pins on 2 channel.

https://youtu.be/_jbd0DPaQg0

I think this driver is very poor for our applications.

temp 80C about 5 minutes work.

r2fpl · « **Reply #177 on:** March 31, 2018, 02:49:19 PM »

I redraw H bridge for Arduino.

Maybe replaced this mosfets to multi mosfet ? I write few post later.

r2fpl · « **Reply #178 on:** March 31, 2018, 04:23:49 PM »

L298n - test 2 channel witch 1 coil

There is still a difference on the channels although it is the same coil.

https://youtu.be/hvrPy_lQfDI

r2fpl · « **Reply #179 on:** March 31, 2018, 06:03:56 PM »

L298n - and another problem.
Overheating 85C ! destability output. We needs better cooling for this driver.

https://youtu.be/TgLBk3GdHSA

News:

User Menu

Author Topic: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated (Read 433033 times)

pmgr

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

e2matrix

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

r2fpl

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

e2matrix

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

listener191

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

listener191

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

cheors

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

r2fpl

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

T-1000

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

pmgr

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

seaad

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

r2fpl

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

r2fpl

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

r2fpl

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

r2fpl

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