Hall sensor question

[fechter]

It's hard to tell from the pictures. The windings look OK to me, just shadow in the pic.

But that doesn't matter to the hall sensor issue. Each sensor signal should toggle as the magnet gaps pass over the sensor. If a sensor is bad, the output doesn't change. If it only toggles sometimes, it would indicate a position problem with respect to the magnets.

 

[larsb]

True.

The relative position of hall and magnets don’t look optimal but in addition to this, isn’t there a damaged wire or some copper visible and scratches on the hall pcb here?



Possibly the red wire by the looks of it, if so the feed to all halls might be intermittently shorted.

 

[Dacflyer]

HI Again... There is no burnt smell in the motor what so ever, and i think things are ok.
I did some more measuring and think if i can get some wider magnets, Things might work.
also i had a discovery..maybe one of y'all can tell me if it's anything meaningable.
this guy gave me all kinds of other things to go with this scooter, and i found an extra hub cover, and This one has some numbers on it.
not sure if it's a model number or just a serial number.. ( XGP2010120012 )

I guess if i get new magnets, I'll probably need to find a solvent to unglue these current ones..
current ones are 1.5" long.

measured the the stater, it is 1.75" wide..
I have room for up to 2" wide magnets.. should i go with the 1.75" or would it be better to go with 2"?
I looked on K&J Magnetics, I don't see what i need, maybe they'll let me know something monday..

 

[Dacflyer]

Larsb >> I looked closer at what you saw, That was just a winding peeking under the hall plate, it's not an issue, Nothing can reach it.
But thanks for the eagle eye...

 

[larsb]

Looks like that wire is being rubbed and the hall pcb too. There will be issues it it isn’t now.

A relation for magnet width in relation to stator width from a science paper i read is to add the magnet thickness on each side so (an example) if you have 100mm stator and 3mm magnet thickness then total width of magnets should be 106.

2” should be about right then.

 

[Dacflyer]

no wires are being rubbed, they was just a bit of winding that didn't get covered in that black goop that covers the halls board..

Ok, I'll go with the 2" wide if i can find them that long.. Stay Tuned....

 

[fechter]

The one picture with the magnet on the stator I'm guessing is not the actual position?

The hall sensors should be all the way in the slot as far as they can go. This will help. The magnets have to pass directly over the middle of the sensor face. If they are right on the edge of the magnets, you could get the kind of goofy behavior you are seeing.

 

[Dacflyer]

fechter >> The halls are all the way in where they are supposed to be.. This is why we think the magnets need to be wider.

 

[Dacflyer]

I just got a reply from K&J Magnetics... this is what they said......

Hi,

Thank you for contacting us.

Unfortunately we don't have those exact sizes. Our BY082 is 2" x 1/2" x 1/8" and our BY0X02 is 2" x 1" x 1/8".

Maybe you could put a BY082 and BY042 side by side to create a 2" x 3/4" x 1/8" magnet. Or two BY062.

Best Regards,

NOW WHAT....Sigh...

 

[fechter]

Do you know the exact size of the original magnets? Most likely a metric size.

There are other places that sell magnets.

You can tile magnets together, but it is extremely challenging to glue them in place because they will repel and want to spread or flip over. I don't recommend that approach. Cutting magnets is not a good option either. I've done both before.

It might be easier to just keep your magnets and machine out the slots for the hall sensors a little so they can be located deeper.

 

[Dacflyer]

Exact size unknown, I thought these were the originals.

Milling the slots longer for the halls might be an idea, but the leads of the halls wouldn't reach them.

 

[fechter]

New hall sensors are cheap.

 

[Dacflyer]

I'll study it, see if it's possible to do without damaging any coils..

 

[larsb]

By looking at this pic it seems like the mid sensor has slid out to the side, this could be at least part of the issue. The other is that the sensors seem to be too deep radially in their cutouts. This way they are more exposed to coil flux and also get weaker field from the rotor magnets. Halls should be on the surface of the steel - as close to rotor magnets as possible.

 

[Dacflyer]

Noted, But i am leaning more to that the magnets are not wide enough..they do seem to be on the narrow side.
Stater is wider than the magnets. I'm guessing that the last owner did stuff that wasn't right.

 

[larsb]

Yes, they are definitely not wide enough, you’ll get 20% more motor with the right magnets. Might as well go for the gold and get N45SH type

 

[Dacflyer]

Larsb >> I'll take whatever i can find that'll fit.. These are N42's.
I've hear that N52 are around as well, But they don't always give benefits.
Now If i can Just find a Source for these longer ones.
Seems like Mission Impossible.

 

[fechter]

The hall sensors being below the stator surface slightly should not be a problem.

 

[larsb]

deep halls is the reason why the hall sensors on my revolt motors gave poor signals. I also have a science paper somewhere where deeper halls was simulated with poor results as coil field interference at high current was higher so i have some support to my post.

 

[fechter]

deep halls is the reason why the hall sensors on my revolt motors gave poor signals. I also have a science paper somewhere where deeper halls was simulated with poor results as coil field interference at high current was higher so i have some support to my post.

OK, I see how that might be an issue at high power levels. It also would depend on the relative position of the magnets. Things get touchy around the edges of the magnets.

 

[Dacflyer]

Ok, I found these, https://www.magnet4less.com/n52-neodymium-magnets-2-in-x-3-4-in-x-1-8-in-ndfeb-rare-earth-block
These are stronger than the N42's Could i use these these or do i need to stick with the N42?

The N42 was listed as 15lb force
The N52 here is listed as 23lb force..

 

[larsb]

Don't get n52, they are expensive for little benefit in motors. Their problem is that their strength is sensitive to temperature and also demagnetise easily in higher temperatures. Normal n42 is not so sensitive to temperatures that way.

A high temp rated magnet can be worth it if you have the funds for something like N42SH but even better N45SH

 

[Dacflyer]

I'm looking and looking, it's not easy to find what i need... looking for 2" x 3/4" x 1/8 1.5" might work as well.. hard to find a source.

 

[Dacflyer]

UPDATE >>> I couldn't find the magnets I needed, so I'm compromising.
I ordered some 1/4 x 3/4 x 1/8
To add onto the edge of current magnets. I'm only doing the 1 side where the hall sensors are.
they are going on ok.. I didn't order enough, 3 short, I wasn't thinking.
but if all goes well, this should work...
any reason it wouldn't work?

 

[amberwolf]

As long as they are the same polarity sequence as the magnets they're on the end of, they will trigger the halls the same way those magnets would.