Author Topic: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated (Read 433596 times)

listener192 · « **Reply #915 on:** May 31, 2018, 04:55:29 PM »

Here is what you get from one set of 5 coils on a 30 slot stator at 25V using the BTN7960 Bridge boards.
When in registration, the inductance of the rotor reduces the current down to about 3A peak. About the same result as the L298N boards.
The BTN7960's don't even get warm.
Increasing the gauge of the wire would reduce coil heat and allow more current at 25V (if required).
L192

Jeg · « **Reply #916 on:** June 01, 2018, 09:51:05 PM »

Quote from: listener192 on May 31, 2018, 02:16:37 PM

A cap of that size will be between 150uF and 330uF.
The voltage rating is that of the cap seen in video 1.I know the size of the, cap as I know the size of the Crydom FWBR along side it.I planned to try 330uF next, unfortunately the relays have failed again so that wont be possible.
The thought was that the half wave ripple may be significant to operation, after all why have such a low value cap there in the first place?
Regards
L192

Hi L192
How much is the positive ripple across the 150uF cap? Is it a 100% ripple or less?

listener192 · « **Reply #917 on:** June 02, 2018, 09:37:02 AM »

Hi Jeg,

The relays failed before I could make any further measurements.
I only have a single bridge board at the moment but with that you can get consistant switching. If you run the clock at 10HZ you can see significant line ripple so I am optimistic about this approach. It will be a few weeks before I get the other bridge boards.

Regards

L192

Jeg · « **Reply #918 on:** June 02, 2018, 10:29:48 AM »

Quote from: listener192 on June 02, 2018, 09:37:02 AM

Hi Jeg,

The relays failed before I could make any further measurements.
I only have a single bridge board at the moment but with that you can get consistent switching. If you run the clock at 10HZ you can see significant line ripple so I am optimistic about this approach. It will be a few weeks before I get the other bridge boards.

First of all thanks for all your kind reports. I had already purchased my relay boards but after yours and Pierre's last failure reports i moved to Mosfets. Just finished my stator these days.

About this "Ripple" approach, i am also positive. But it will need a some kind of sync between switching and ripple's minimums/maximums.

listener192 · « **Reply #919 on:** June 02, 2018, 12:57:55 PM »

Hi Jeg,

The only hole in this theory is that the system still functions off the super caps, although this could just be less than OU as observed. Why have a frequency control at all? Perhaps it is really being used as a crude phase control when running the transformer from the line supply. When running looped the phase will be the same.

Regards

L192

Jeg · « **Reply #920 on:** June 02, 2018, 05:21:09 PM »

Quote from: listener192 on June 02, 2018, 12:57:55 PM

Hi Jeg,

The only hole in this theory is that the system still functions off the super caps, although this could just be less than OU as observed. Why have a frequency control at all? Perhaps it is really being used as a crude phase control when running the transformer from the line supply. When running looped the phase will be the same.

Well, looks like that caps are just for absorbing the spikes. It should be less than OU as "Spike return" never showed something promising. At least by itself.

I don't really see why the phase will be the same when running lopped. As i am thinking it, a maximum value of ripple voltage (120Hz), should be presented when we have a full registration as you call it during both cycles.

Some pics for the records

pedro1 · « **Reply #921 on:** June 02, 2018, 10:59:48 PM »

J'ai voulue faire cette méthode mais je me suis rendue compte a mi-chemin que cela ne fonctionnera pas très beau travail quand même et cela pourra donner un excellent transfo pour le reste j'ai quelque reticence

En. I wanted to use this method but I realized half way through the build it won't work as I thought but nice job anyways and it could make a great transformer! but for the rest I have some reservations.

Jeg · « **Reply #922 on:** June 03, 2018, 12:13:27 AM »

Hello Pierre

Thanks for your comments. Yes it is a different approach and it will not work by driving it the same way as your first prototype.

Regards
Jeg

pmgr · « **Reply #923 on:** June 03, 2018, 07:07:03 AM »

L192 et al,

With respect to the diodes and how Pierre has them hooked up, please see the attached images (I note that Pierre does not have the ground side diodes).

In the attached images, Vdd is connected to the rectifier bridge (small red cable that drives the relays). Vc is connected to the capacitor bank.

Since Vdd is always larger than Vc, the diode (D1_p in the attached image) that is connected to the relay that is turned on (S1_p in attached image) will always sink current to the capacitor bank through that relay.

So if the capacitor bank is not fully charged (or empty), it will be mostly charged directly through the relay and the diode and not through the 4 ohms resistor.

So it is very important that the capacitor bank is fully charged up to Vdd before starting the Arduino, otherwise it will be very likely that too much current will go through the relay (and diode) and overload the relay and maybe even the diode if its current rating is too low. The 4 ohm resistor is not going help much (if at all) in limiting the current to the capacitor bank in this case as it is parallel with the relay/diode.

And once one relays fails, the other switched on relays are going to get the current of this relay in addition to what they are already carrying and so they will fail as well. And since the Arduino is switching all relays one after another, eventually all relays will quickly fail.

And I note that this will also be the case if you are using H-bridges or MOSFETs. A capacitor bank can sink many many amps of current, so it is crucial to put a fuse in the main Vdd line, between rectifier bridge and relays, and make sure the capacitor bank is fully charged up.

Pierre, this is probably what caused all your relays to burn out.

There is a simple solution to this problem: make sure the capacitor bank is fully charged before starting the Arduino and use a fuse.

Lastly I note that if your output coil is sinking too much current and draining the capacitor bank to a voltage too low, this problem will also occur once the capacitor bank gets discharged to a value a lot lower than Vdd.

PmgR
====
Help end the persecution of Falun Gong * www.faluninfo.net * www.stoporganharvesting.org

r2fpl · « **Reply #924 on:** June 03, 2018, 08:20:29 AM »

Quote from: pmgr on June 03, 2018, 07:07:03 AM

L192 et al,

With respect to the diodes and how Pierre has them hooked up, please see the attached images (I note that Pierre does not have the ground side diodes).

In the attached images, Vdd is connected to the rectifier bridge (small red cable that drives the relays). Vc is connected to the capacitor bank.

Since Vdd is always larger than Vc, the diode (D1_p in the attached image) that is connected to the relay that is turned on (S1_p in attached image) will always sink current to the capacitor bank through that relay.

So if the capacitor bank is not fully charged (or empty), it will be mostly charged directly through the relay and the diode and not through the 4 ohms resistor.

So it is very important that the capacitor bank is fully charged up to Vdd before starting the Arduino, otherwise it will be very likely that too much current will go through the relay (and diode) and overload the relay and maybe even the diode if its current rating is too low. The 4 ohm resistor is not going help much (if at all) in limiting the current to the capacitor bank in this case as it is parallel with the relay/diode.

And once one relays fails, the other switched on relays are going to get the current of this relay in addition to what they are already carrying and so they will fail as well. And since the Arduino is switching all relays one after another, eventually all relays will quickly fail.

And I note that this will also be the case if you are using H-bridges or MOSFETs. A capacitor bank can sink many many amps of current, so it is crucial to put a fuse in the main Vdd line, between rectifier bridge and relays, and make sure the capacitor bank is fully charged up.

Pierre, this is probably what caused all your relays to burn out.

There is a simple solution to this problem: make sure the capacitor bank is fully charged before starting the Arduino and use a fuse.

Lastly I note that if your output coil is sinking too much current and draining the capacitor bank to a voltage too low, this problem will also occur once the capacitor bank gets discharged to a value a lot lower than Vdd.

PmgR
====
Help end the persecution of Falun Gong * www.faluninfo.net * www.stoporganharvesting.org

PmgR

Yes: 4 ohms they depend on the voltage on which the voltage capacitors are. It allows current from either the bridge or capacitors and it is important to fully charge the capacitors before switching on.

T-1000 · « **Reply #925 on:** June 03, 2018, 10:09:59 AM »

Quote from: Jeg on June 02, 2018, 05:21:09 PM

Some pics for the records

Hi Jeg,

The magnetic field polarities will be on wrong direction with such coil arrrangement. Please see https://www.miniphysics.com/ss-magnetic-field-due-to-current-in-a-solenoid.html for explanation.

Cheers!

listener192 · « **Reply #926 on:** June 03, 2018, 10:37:34 AM »

Hi PmgR,

Good point. If the presence of line ripple is not a significant factor then suggest using a switch mode power supply with current limiting is used and leave out the 4 ohm resistor. This way initial charging current is limited and just use the super cap bank for supply and recovery.

With the new bridge board, out of registration the current draw was 6A peak from a 25V supply.so 18A with 3 coil sets on, closer to 36A with overlap coils on. This was measured at 10Hz clock. So even with a 40A current limit this should be within the rating of the 45A rating of the bridge board.

When in registration the inductance increases and a single coil set current draw drops to 3A or 18 A in total. As the out of registration coils perform no useful work, the current limit could be set to 20A. If the recovery voltage rises above the current limited supply, then more current could be drawn by the bridge boards without limiting.

Regards

L192

seaad · « **Reply #927 on:** June 03, 2018, 01:13:22 PM »

Quote from: T-1000 on June 03, 2018, 10:09:59 AM

Hi Jeg,
The magnetic field polarities will be on wrong direction with such coil arrrangement. Please see https://www.miniphysics.com/ss-magnetic-field-due-to-current-in-a-solenoid.html for explanation.

possibilitys

Reagards Arne

Jeg · « **Reply #928 on:** June 03, 2018, 02:48:57 PM »

Pmgr
Thanks a lot for the analysis.

T-1000
I don't understand your point. Do you mean that torroid cores and normal winding coils can not produce OU as Akula and so many others did many times now? Anyway, i know that i move to something different than what Pierre showed but i have something very specific in my mind to try.

T-1000 · « **Reply #929 on:** June 03, 2018, 03:36:30 PM »

Quote from: Jeg on June 03, 2018, 02:48:57 PM

T-1000
I don't understand your point. Do you mean that torroid cores and normal winding coils can not produce OU as Akula and so many others did many times now? Anyway, i know that i move to something different than what Pierre showed but i have something very specific in my mind to try.

When coils are like in Pierre's contraption they are facing about 20 degrees off the pickup coil core. In your photo they are on 90 degrees. So the magnetic polarities as they produce magnetic lines towards closest opposite pole.
And in regards to Pierre's case it is not just single coil on core, the multiple coils are sharing same core and are positioned with offset to the side on each which allow different 3D location of magnetic poles towards center on each coil powered on.

News:

User Menu

Author Topic: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated (Read 433596 times)

listener192

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

Jeg

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

listener192

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

Jeg

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

listener192

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

Jeg

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

pedro1

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

Jeg

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

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

listener192

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

Jeg

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