Hi Gyula


As you can see your no hall mosfet pulse motor circuit suggestion works very good, so thanks again. I just needed to tweak some components with different values.
I will try mex circuit and see which works better for BEMF recovery, but I need to order P channel mosfets first.


Just one question for you Gyula, will mosfets in mex circuit survive 270volts spikes from drive coil or I need higher voltage rating mosfets?


JoeFR

Quote from: mariuscivic on November 21, 2011, 09:01:14 AM
Hi Joefr!

This is how i connected everything. One question: is it posible feeding the first driving circuit with the output from the leds without shorting or burning anything?
Thanks!

Hi Marius,

The answer to your question is the same I gave to Romero back then: you need to use a DC/DC converter (a switch-mode power supply with isolated input and output grounds).  Such converters insure a stabilized DC output voltage which is important if you wish to avoid a run-away situation: without a stabilized output voltage the full setup can distroy itself very easily, unless you are fast to use a 'safety switch' built in advance to interrupt current the moment a run-away starts developing.

On isolated input and output grounds I mean that there is no galvanic connection between say the negative input and output pins of the DC/DC converter.  This is a must in your case because your puffer capacitors within each stage do not share a common connection (either positive or negative) with each other, so an isolated converter insures the proper operation (it prevents the short circuits between circuit points that should not be connected but they do get unwantedly connected when no isolation means is introduced).
Unfortunately, the number of 'isolating solutions' depends on how many stages you wish to run: at the moment you have two stages so one 'isolating means' in the form of such a converter is needed. If you use 3 stages than two 'isolating means' are needed but the second 'isolating means' need not be a DC/DC converter like the first one: it could be an 1:1 AC transformer for instance (able to work as a normal transformer at the rotor's rpm frequency), connected in parallel with the second driver coil, this would give the galvanic isolation.  Of course the AC transformer could be substituted by the driver coil you make with two parallel guided wire instead of a single wire, to get the 1:1 transformer in the driver coil itself.  Tomorrow I can draw how I mean if this is not clear yet.

I would invite everybody here to solve this 'isolating and stabilizing problem', including our new member, mex too.

However, I believe you need to build some more driving stages than just the two you have showed because probably this present two driving stages would not be enough as yet to maintain a looped situation,  I just guessing here now. 

rgds,  Gyula

Hi Gyula!

Muller is not exactly the device I made the circuit, other settings I'm using magnets, coils and other seeds as well.

Hi Gyula!
This is the circuit I'm using it right back so I can control the energy of the BEMF 4700uF capacitors. An oscilloscope can be easily measured in a series of low-value resistor, the voltage, the BEMF during the period covered, and the calculated efficiency. RESISTOR and D4 of the anode must be connected between the negative pole.
The coils used depends on the setting of the potentiometer can be adjusted.


joefr!

BEMF will not be large, because the capacitor absorbs enough of the IRF and 9530 as we

Hi Joe,

Pleased the circuit operates well. 
Regarding the IRF9530 p channel and IRF530 n-channel types in mex's schematics, they are both 100V devices only, and while they are in series connection, the combined 200V reverse voltage sounds less than your 270V recovered voltage.    BUT  mex's circuit directly connects the recovered energy in parallel with the input DC voltage and IF there is no any external DC supply input, the circuit generates it when the rotor is speeded up.  If there is a battery as an starting input supply in parallel with the 4700uF cap, then the low inner impedance of this battery keeps the peak voltages at low level as a load that is being charged but in case there is no battery then the DC voltage can go up to pretty high so a load should be used, otherwise the BC328, BC338 transistors would be damaged if you would let the voltage in the puffer cap go up to above 30-40V in the unloaded case.
So this is a bit different circuit from the one you use to get the 270V DC in your 660uF cap, ok?
(EDIT: maybe I am mistaken and you do not get such a high DC in the 660uF cap, only the spikes have 270V peak values, ok? )

Otherwise You can use your 300V p-channel type you used earlier  (you uploaded it in the previous page I think) and also you need at least a 300V n-channel MOSFET too, BUT then the original FET types are good when an input battery is used to run the setup and being charged from the captured energy.

Gyula

Quote from: joefr on November 21, 2011, 05:14:00 PM
Hi Gyula


As you can see your no hall mosfet pulse motor circuit suggestion works very good, so thanks again. I just needed to tweak some components with different values.
I will try mex circuit and see which works better for BEMF recovery, but I need to order P channel mosfets first.


Just one question for you Gyula, will mosfets in mex circuit survive 270volts spikes from drive coil or I need higher voltage rating mosfets?

JoeFR