DIY Electric Car Forums banner

1 - 7 of 7 Posts

·
Registered
Joined
·
17 Posts
Discussion Starter #1
Hi everyone,

We recently got hold of a nice HVH250 (without the casing, and that's a PITA, but it's another story...) and we're planning to power it with a nice 400V/400A controller.

Except... that the controller only accepts hall-sensors and not cos/sine resolvers like the one that is mounted on the HVH.

Do you guys know of a ready-made solution to convert the cos/sine resolver signal to a classic UVW hall-effect sensor signal?

I know IChaus has some nice integrated solution like this one :
https://www.ichaus.de/upload/pdf/MH8_datasheet_A1en.pdf

but it does involve integrating a small magnet and more incidentaly, it says it can only handle motors up to 8 poles (4 pole pairs) and the HVH250 is 14 or 10.


Thanks for the feedback!



Marc
 

·
Registered
Joined
·
17 Posts
Discussion Starter #3
Hi Tomdb,

Thanks for the link, it is very interesting!

I would have loved to have just a tad more precision (360/1024 gives more or less 0,3°) to get to the 4096 steps on a revolution that is the state of the art. After all the resolver is there to give the most precise information possible (apart from it's ruggedness).

But maybe I'm splitting hairs on this one... I welcome any feedback on this!

A couple of questions on this design if some of you are interested enough to go and take a look at it :

  • It states a 2kHz sine excitation for the resolver, I can't find any information on the HVH250 excitation frequency / amplitude. How would I know if that would be ok for it?
  • The HVH250 I have is a 10poles motor, what should I put as commutation angles? 360/10 = 36°. So that means that very other 36°, either U or V or W (hall sensors information) switch to on or off? Is that it?
    Also, how would I know if I'm correctly angularly aligned with the real commutation angles?


I'm still getting the grasp of basic BLDC control so thanks for your help!


Marc
 

·
Registered
Joined
·
934 Posts
Using the code as a base you could very easily use a more capable micro to boost the measuring ferquency. For example a teensy 3.1 https://www.pjrc.com/teensy/teensy31.html.
however, since you just have 10 poles, 5 pole pairs, Hall sensors just need to update roughly at 36 degrees per revolution.

So without knowing what controller you got 0.3 degrees sound very accurate for rotor position using just hall sensors. Mind posting what controller you are think of?
 

·
Registered
Joined
·
17 Posts
Discussion Starter #5

·
Registered
Joined
·
934 Posts
Got any further details on the controller? I cannot derive what kind of algorithms or output wave form is generated.

Because maybe it can drive the Remy in a sensorless mode, not optimal but it can work.
 

·
Registered
Joined
·
1,478 Posts
i think you can determine the commutation sequence and timing by back-driving the motor with another smaller motor and looking at the bemf waveforms on a scope. i have done this on a big cnc lathe motor which used a 6-phase tachometer for commutation and had to be aligned to the motor windings.

Analog devices makes a resolver to digital converter hybrid that you could manipulate the output in microcontroller code to create the hall signals. i think it generates the 2 or 4 khz excitation also.
 
1 - 7 of 7 Posts
Top