Muller Dynamo

[teslaalset]

What is missing in the alternative circuit is a decent flyback diode.
If you skip that, your transistor likely will die from overvoltage caused by the voltage spike when the transistor is switched off.

[e2matrix]

I was talking about the resistor from few posts before where I can see a resistor of 10k to the base of the transistor.
I was not using that, i had a resistor on the input to the hall but yes u can have a resistor to the base too but I think 10k is too low.I was suggested 20k to the base.I have never tested it like that.
yes my input resistor was 100ohm

Thanks Romero.  I had not seen the reference to 10K (just found it now in ChricC's diagram - getting up to speed here slow but sure   ) but saw Gyula mention 470 Ω to 1.5K Ω.  Not to keep bothering you with all the little details I'll ask anyone else that's clear on this if the pic below looks correct for what we are talking about.

i_ron I see we were posting at the same time - thanks for the suggestions.  At this point I just need to know the details of a circuit that will work.  I'm not smart enough to build a circuit without values on everything.  I've got everything for the circuit below so if that will at least get things going I'll go with that for now and work toward a much more efficient and less likely to smoke circuit later. 

[i_ron]

What is missing in the alternative circuit is a decent flyback diode.
If you skip that, your transistor likely will die from overvoltage caused by the voltage spike when the transistor is switched off.

It is not a transistor, it is a FET

The Fet has a body diode that, at these levels, works adequately, but yes
I was presenting as simple a concept as possible, feel free to add an additional diode. Nor have I shown the necessary decoupling caps on the Fet driver.

Kind regards,
Ron

[gyulasun]

Hi Folks,

I included below a modified schematic of Romero, and I inserted a series resistor between Hall Pin 3 and transistor base to limit the collector current of the Hall's output transistor and the base current of the PNP transistor.

The total peak current is limited only by the two series driver coils when the transistor is ON, if the self inductance of someone's coils is lower than that of Romero (about 2 x 1.2mH) or someone's switch ON time is longer than that of Romero used, then much higher current can flow via the Hall output and the base-emitter junction of the PNP so that heat can develope, either the Hall or the PNP transistor can get ruined as freenrgy850 reported the latter. Woopy somewhere wrote he already inserted a resistor as per the schematic below and he found current draw decreased, hence input power need also went down. But he used the original Romero circuit earlier without problem if I recall correctly.

Just noticed e2matrix also uploaded Groundloop nice schematic with a 10-20kOhm suggestion, it may prove to be too high in value nevertheless it can be tested.  This value greatly depends on the DC Beta (current amplification factor) of the PNP transistor (dI_c/dI_b).

Please understand:  the circuit as Romero showed without the resistor has been working for him without any problem and works for others too, I included the two main factors that may lead to overheating if exceeded.
For someone wishing to use FET driver ICs as manufacturer Application Notes show them or recommend them, it is also ok, mainly for long time operation, but for testing purposes Romero original circuit is just fine but you may wish to observe the recommandations above.

rgds,  Gyula

[i_ron]

I'm not smart enough to build a circuit without values on everything. 

Sorry, I posted this as an idea piece at the beginning of this thread. I forget who, corrected my sketch and added the values. So I was assuming people had seen it.

But on the spec sheet it calls out 10k to pin three (R2) so anywhere from 6k to 10 k is fine. The gate protection resistor (R1) is again, just anything from 10 ohms to 50 ohms, I often use 4.7 ohms as I have a bag full.

Most fet drivers like 1mfd right at the supply pins, again all rule of thumb.

Ron