encmode ABZ [SOLVED]

[GrapeNuts]

I'm having an issue when encmode is set to ABZ my motor will not spin when I start the inverter. If I spin the motor shaft by hand I do see the speed value is registered by the inverter's plot function.

With all other parameters the same if I change the encmode to AB the motor will spin when I start the inverter. I'm confused by this behavior. Any suggestions for what may be causing this?

[arber333]

You use PMSM motor or ACIM? If you use ACIM with ABZ it will not work. Use AB!

[GrapeNuts]

Unfortunately encmode AB doesn't seem to work very well for me either.

The motor is an ACIM motor and it is equipped with a dynapar incremental encoder. The encoder provides an ABZ signal. The model # is HA625102403EA. From what I can tell reading the datasheet at https://ecatalog.dynapar.com/ecatalog/i ... rs/en/HA25 the device provides 1024 ppr.

For some reason when I configure my inverter with these options the inverter will not run with encmode set to ABZ and with encmode set to AB the speed plot is very erratic, but the motor does run for a short time.

[arber333]

Hi
You should use AB channels with this motor. Leave Z signal unconnected.
1024 pulses is a lot. Does it have 1024 each channel?
Do you power it from 3V3 or from 5V? Maybe it needs higher voltage like 12V?

I have also seen encoders that signal in rising edge. It is a sort of shark fin and it rises with rotation...
Can you connect scope on channel A and rotate motor manualy? You should either see pulses in 5V if signal is push-pull or 1.6V if signal is open collector type.

EDIT: I googled your link and i see you use 1024ppr with 5V to 26V line driver. That means you should see a 5V signal. Probably your board is supplying only 3V3 and it is not enough. Try connecting 5V or more to encoder supply. I use 5V. But i used 12V directly from a car with open collector encoder, and it worked. You should proably use stabilised voltage though...

[GrapeNuts]

Yes I believe it provides 1024 pulses on each channel. This image is taken from the documentation.



I attempted to power the encoder with the 5v provided from the ENC_5v pin, but it seems the encoder draws too much current and the voltage sags to 2.1v and the encoder has no output.

When I power the encoder with the 12v source the output pulse is 11.5v.

[johu]

You might want to replace R7 by a bridge. It seems to draw 25mA causing the drop from 5V to 2.1V.
The pulse count is quite beyond the design limits but you can adjust this replacing R2 and R5 by 1k. The digital filtering within the processor might still pose a limit but you should be able so spin up to a certain speed.
Like Arber said, omit the Z channel, I don't know why they have it on an induction motor.

[arber333]

Yes I believe it provides 1024 pulses on each channel. This image is taken from the documentation.



I attempted to power the encoder with the 5v provided from the ENC_5v pin, but it seems the encoder draws too much current and the voltage sags to 2.1v and the encoder has no output.

When I power the encoder with the 12v source the output pulse is 11.5v.

You have to be carefull not to overload the A and B channel inputs. Since you push 11V INTO them you should consider the resistor voltage divider schematic of the A and B inputs. You may want to change those resistor values...
Open collector outputs do not have that issue.

[GrapeNuts]

Thanks johu, arber333. I will remove the Z signal and try the resistor value modifications you have suggested.

[GrapeNuts]

Obviously the ppr provided by my encoder is too high. Its been a long time since I was in college and I only have a vague recollection of my digital logic course, BUT I do remember discussion of frequency division techniques. The trick now will be trying to remember exactly how they worked. I think using a decade counter.

I may try to reduce the ppr with such a technique. What is considered an ideal ppr rate for an induction motor?

[arber333]

I considered using 256ppr but then I made a ring with 60 holes and optic sensor A and surprisingly it worked good. Now I use AB magnetic pickup sensor with 64 teeth ring on the rear side of motor. I drive every day and I dont think about sensor any more. Like a friend said to me: if you find spiders nesting in a box then it must be a good design, since nobody disturbed spiders!

See the back side of the motor with encoder and rpm sensors.
https://mazdamx3ev.wordpress.com/2014/0 ... indikator/

[bundao]

Just a point of clarification on the voltages of the AB encoder signals on the V2.0 board. I noticed the inputs being used (PA6, PA7) don’t seem to be 5V tolerant, yet it seems to me that they would be subject to 5V even at switch on... but I'm not an electronics engineer.
I’m using an open collector type encoder and it works fine, but I know another user who is going to use a resolver to encoder board and we are wondering what precautions to take regarding the signal level.
Any help/suggestions much appreciated.

[johu]

There is a series resistance of 10k on each channel. In the manual it states that you must limit the current into none-5V-tolerant pins to a couple of mA. With those series resistors that constraint is met.

[bundao]

OK so the voltage can be higher than 3.3 as long as the current is kept below the specified value.
Thanks for the clarification.