help with IR2110 chips

[Alec_t]

Experiment time.

 

[Harald Kapp]

as you can see there is a phase shift of 90 degrees on each coil

Yes, now that you changed the parameters of the voltage sources...

 

[Maglatron]

so do you think that it would be better with 18 on each side then or 16?

Experiment time.

 

[Maglatron]

I'm going with 18 final answer

 

[Maglatron]

should get chance to stick the magnets today, got the glue!

 

[Alec_t]

Having more magnets will increase the frequency at which each coil gets energised, so coil impedance due to inductance will increase too. Have you considered the effects of that? From your pics it looks like the coils are quite bulky and unlike the air-cored pancake ones often seen on axial flux generators. Their inductance is likely to be much higher than the pancake type, especially as your coils have a ferromagnetic core.

 

[Martaine2005]

If I were replying to this thread, I’d stick to ‘experiment time’.

 

[Maglatron]

so I had the time to mark out and stick the fisrt layer of the first side took half hour, hopefully get the rest stuck tomorrow and I'll stick the rest but you were dead right on the 18 magnets fit like a dream, looks so damn nice! could be art lol, as soon as the magnets have been stuck and dried I'll put the wheel on the machine and run some experiments on the fitted coil then procced to fit more coils!

Having more magnets will increase the frequency at which each coil gets energised, so coil impedance due to inductance will increase too. Have you considered the effects of that? From your pics it looks like the coils are quite bulky and unlike the air-cored pancake ones often seen on axial flux generators. Their inductance is likely to be much higher than the pancake type, especially as your coils have a ferromagnetic core.

I had considered some of that, and when I say considered I tought about it for a minute but didn't do any of the math. the coils turn out to be 5cm in diameter and about 7.5cm long and do have ferrite core (steel bolts) I'm exited to see the out come difference on the one coil that is fitted, I removed the 50 ohm coils on the back, both of them but like you said the coils are quite bulky but I like them, do you know where to source some 0.7mm 500grams spools of mag wire, I chose an uncommon size! wasn't really thinking it looked and sounded good but now I can't find any that that have the same dimensions that ship before march the 4th so any pointers will be welcomed on that front but all in all happy good day! hopefully have some pictures for you soon! thanks
experiment time!!!

 

[Maglatron]

I'm going on the hope and whim that the flywheel effect will iron out the bumps to a good enough degree only one way to find out eh!

 

[Maglatron]

sticking the rest of the magnets today!! hopefully can take a photo or two!

 

[Alec_t]

do you know where to source some 0.7mm 500grams spools of mag wire

I don't even know which country you're in. I'm sure Google will know a likely source of wire.

 

[Maglatron]

I'm in England Mr Alec_t

 

[Alec_t]

Having more magnets will increase the frequency at which each coil gets energised, so coil impedance due to inductance will increase too.

On second thoughts, for a given wheel rotational frequency the coil output frequency per energisation pulse is dependent on the magnet size relative to the rotational circumference, so impedance should be relatively independent of the number of magnets.
Using this calculator, your present coil's guestimated inductance seems undesirably high.

 

[Maglatron]

I get to work on it for an hour and a half today so I will see the outcome of the coil and I'll mount the second that I have wound and take photos see what you think

 

[Maglatron]

I didn't count the turns

 

[Alec_t]

Assuming your 5cm diam coil is wound on a 1cm diam core, is 7.5 cm long and wound with 0.7mm diam wire you should get about 100 turns per layer and about 14 layers. Assume core relative permeability is in the 1000-3000 region for layers near the core but about 1 for layers near the outside of the coil. Assume average turn radius is 1.5 cm.
That should give you a rough idea of inductance when using that calculator.

Edit
For '14' read '28' .

 

[Maglatron]

cool!! would it be useful to take an average of 1 + 2000 / 2 = 1000? run out of glue!! piss take, nevermind, another day another dollar

 

[Alec_t]

would it be useful to take an average of 1 + 2000 / 2 = 1000?

Depends how the influence of a high permeability small diameter core varies with the radius of a bigger turn. Haven't researched that, but gut feeling tells me the influence drops off rapidly with distance. For my rough guestimate I assumed only 1 layer had u=2000 and 13 layes had u=1. That gave an inductance value of at least 50mH as I recall. Even that will drop a considerable voltage.

 

[Maglatron]

I don't think that that calculator is good I need the core to be a centimeter and the daimeter of the coil to be 5 cm so thats 2cm of coil either side of the core I think what you said about the length of the coil and the turns was clever because I do think there are about 100 turns / 7cm of core I think that there are alot more than 14 layers I think there are 28 layers = (20mm / 0.7mm) there is also the fact that the coils wiegh 500 grams and the diameter of the wire is 0.7mm can you work out the lengh with this information along with the density of the copper maybe? and calculate the turns with that but from the 100 * 28m that should be around 2800 turns which turns out to be quite pheasible saying that they take an hour to wind!

 

[Maglatron]

Depends how the influence of a high permeability small diameter core varies with the radius of a bigger turn. Haven't researched that, but gut feeling tells me the influence drops off rapidly with distance. For my rough guestimate I assumed only 1 layer had u=2000 and 13 layes had u=1. That gave an inductance value of at least 50mH as I recall. Even that will drop a considerable voltage.

drop??