Recent Posts

Pages: [1] 2 3 ... 10
1
Very nice! 

No, no swapping of wires is necessary.  Under the encoder section, it has a swap direction setting.

Our encoders too do not tolerate much radial or axial movement.

Tip:  Shim the axial runout of your rotor so it is under .010"  so encoders will work.

Last tip: Encoders will NOT work with the outer cover loose or off. 

After you afix your encoder permanently, run the motor characterization routine. (to get your motor to run perfect.)

Grats    :co<:
2
PULL-UP Resistors Not Needed.

I received the eBay encoder today.
It is Chinese made (of course for $19).
I was not disappointed with the feel. It is a little larger than I expected. It does appear to have a real bearing, not a brass oil-lite slip ring.
It has small screws that allow the protective cover to be removed. I may take it apart this weekend to get a better look at the construction.
I am certain that this unit will not tolerate any axial loading so final mounting must be precise and square to armature or longevity will diminished.
For the price I cannot complain.

More importantly after some tests I can confirm that pull-up resistors are not needed with this encoder and pull-up is being performed inside the controller. This makes interfacing much simpler and cleaner.

I hooked it up and it does work. My thin (testing) 3 phase cables cables run very cool now that phasing is locked to armature position.
I fabricated a temporary rubber bushing to insert into the armature shaft for testing.
I slowly brought the motor up to 900 rpm and I am sure I could have went much higher but I did not want the encoder to go flying off the end of the motor shaft as it was just sitting there with friction. I will go faster once I fabricate a better mount for it. I ramped the motor up slowly and did not see it spike above 3.5 amps at 24 volts. (unloaded motor)
I did offer some resistance to the motor and I did observe that the controller maintains speed and current climbs accordingly.
This encoder is 100 pulses per revolution so you must adjust the controller ppr value to 100.
If you order one of these make sure you do not get a version with higher PPR as the controller will not accept any high PPM values over 256. A lot of the units I saw on eBay were 300 ppm or higher.
The encoder will work with the regulated 5 volt supply from the Curtis controller but can accept any voltage up to 24 volts.
Red to 5v, Black to I/O GND, Green to PhaseIn 1 and White to PhaseIn2. Two phase wires may have to be swapped depending on orientation.
I am posting this here just in case someone else here wants a low cost solution for their project.





Thanks
3
WOW! You are worse than me....LOL

Lets see, projects for the last year or so....

Repowering a Chinese 2 person go kart.
Building and testing about 15 WiFi antennas.
Building a new motor for my truck.
Clearing out old half finished projects like: a powered red wagon,  self built Segway (Because I could not debug code)
Things still pending: 2 electric guitars (in pieces),  2 broken amplifiers.
Not to mention: Maintenance on my electric Roadster.

4
@mizlplix

The controller was a gift from a co-worker. They were pulled from some forklift gear at the request of the OEM and were shelved. They were basically new. I don't know why they were pulled. Maybe they had to re-rate the carriage capacity after some aftermaket mod and add a bigger controller.
When it was time to clean up they discarded them. I was fortunate to get one.
The motor was also pulled from a forklift. It was pulled due to the bad bearing-encoder. It is not field replaceable so they replaced they entire motor. I got the bad one for free.

So far I have less than 30 bucks into this thing and I plan on keeping it low. Even the programmer was done via barter. I just had to solder 2 opto isolators onto the board and it was ready to go.

The motor and this controller are a good match.

The ebay encoder is dual channel. No problem there. The power supply is a temporary solution to see if I can get the motor to spin. I have no illusions that a bench top power supply of 30 amps will produce any useful work but it will spin a motor up if it is ramped up slowly. I have built homebrew AC motor controllers from scratch so I know the demands are high. Some of those adventures can be seen here: https://youtu.be/zpJWGD5JtLA.  My homebrew controller did not require two channels of encoding so the bad encoder bearing with one working channel would have been fine. My motor was never to spin backwards. I sold that controller to a chap in Cali when I changed direction on building a full electric car. After doing the math it was cheaper for me to just buy an i3. With that I got a rocket with full climate control, integrated infotainment, no ECU problems, canbus problems, airbags, etc.

So the motors I have are the leftovers from another project. They have just been sitting in my garage. I want to make a nutty go cart or something but I don't want to get crazy spending money on it. I would rather take that money and invest in Apple and Intel stock :). But I do like to tinker so why not try? If I get sidetracked I can always sell the controller and motor right?

If I can get this motor and controller to cooperate then maybe I will be willing to invest in some lipo battery tech and weld up an nice tube frame.
This project is a little more fun than the alternative which is me modifying and old BigTrak toy from the 80's and adding some intelligence to it :) I can't sit in a BigTrak though!
God, I have to get a life  :sf<:
5
There is a setting in the controller to set the encoder count.  It comes set at 64 pulses.

It is also dual channel.  It reads both. 

If you can not get the pulse count set to match or if you do not have dual channels in the encoder, it will not work.

Is there any good reason to mess around with the wrong motor, the wrong encoder and the wrong power supply?

Assume it all works from the factory. Then set it up on the intended motor.  You will have a lot less headaches!
6
@Ivan

This is what I purchased. I think it is similar in overall specs. Probably not as robust. I do not plan on using this in an over-the-road application. I don't see a need for an index channel so I should be OK there. It has a wide operating range so I think I am ok on the TTL front. I will be using the 5 volts line from the Curtis unit. The Curtis manual makes no mention of whether they internally use pull-up resistors though. They don't draw them on their mock-up diagram either.



New Encoder 100 P / R 5V-24V Incremental Rotary AB 2 Phase 6mm Shaft

Features :

100 p/r (Single-phase 100 pulses /R,Two phase 4 frequency doubling to 440 pulses)

Power source: DC5-24V

Shaft: 6*13mm/0.23*0.51"

Size: 38*35.5mm/1.49*1.39"

Output :AB 2phase output rectangular orthogonal pulse circuit, the output for the NPN open collector output type

Maximum mechanical speed: 5000 R / min

Response frequency: 0-20KHz

Cable length: 1.5 meter

Notice:AB 2phase output must not be directly connected with VCC!!
Connection:

Green = A phase, white = B phase, red = Vcc power +, black = V0

Application:

Is used to measure the rotational speed, Angle and acceleration of the object and the length measurement

Suitable for intelligent control of various displacement measurement, automatic fixed-length leather automatic guillotine machines, steel cut length control, civil measured height human scale, Students racing robots.
7
This is the encoder that works with the curtis controller.
Does your ebay encoder have the same specs?   :mA<:

https://www.usdigital.com/products/encoders/incremental/rotary/kit/E3

64 to 10,000 cycles per revolution (CPR)
256 to 40,000 pulses per revolution (PPR)
2 channel quadrature TTL squarewave outputs

Pins on encoder
1   Ground  --  NO Connections
2   Index
3   A channel
4   +5VDC power
5   B channel

Ivan
8
I hooked the controller up to a larger motor this weekend and I did get smooth response but not beyond low speeds.
The motor I have has a SKF main bearing with a BAD built in encoder (hence why it was free). I am forced to use an external encoder or pay for the ridiculous cost of a new SKF bearing. For me it is more logical to put the armature on a lathe, drill a hole and tack weld a D shaft to protrude from the end bell to interface with an easy to access encoder.

No encoder signal coupled with the fact that I am running the entire thing of a low current bench supply explains the low RPM's.
The original encoder was 80 ppr. I bought a 100 ppr encoder from ebay to continue my experiments. It should arrive any day :)

When I looked up the original-bearing encoder data sheet they have pull-up resistors between VCC and each phase. The pull-up values are swapped out depending on what the VCC voltage is.

My question is: Do I have to do something similar with the new encoder I am getting? Or are the pull-ups already inside the Curtis controller? Can I just connect the four wires of the encoder directly to the controller with no fear of damaging anything??
9
When you reflashed the firmware, you removed the HPEV firmware.  So, all of the "motor type" files were flushed.
Those were for HPEV  made motors only so your little demo motor would not be there any ways.

Having said that, a perm mag motor does not have the same rotor mag configuration as an induction motor.  It would drive the Curtis controller crazy because the adaptive strategy is induction motor only.  BUT, (Just a wild idea) you could mount an encoder to the little motor,  it just might work......might and might not.  Only a Curtis engineer could tell you.  Could you sent me a picture of your test motor? 

Sorry about the late help, I'm 66 and I have my good days and not so good days.  I have been indoors for the last few days....

Glad you got the controller to at least twitch a small motor.

NOW:  After a new reflash or change to a new motor, you will need to hook up to the new motor and run the "Motor characteration " routine so it can do all of it's internal settings before the controller can really take advantage of it's new parameters.

Set the test amps to 300.  Then follow the instructions here: http://ivanbennett.com/forum/index.php?topic=57.0

NOTE:  Your "motor type" parameter will always read as "0".  as all the files were flushed.  Changing this after doing the motor characterization routine will cause you to NEED to do the routing over again!

Let us know.....
10
When you flash a controller    :Ff<:   there is no motor type that was for hpev (VCL) different motors.

TYPE 50 for the AC-50  ect....

Your type should be on 0  zero

The curtis controller has to have an encoder, think of it this way, how does the controller
know which way the motor is turning? CW or CCW or the RPM     Make sense

Pages: [1] 2 3 ... 10
Image Ad - Leaderboard

Image Ad - Leaderboard
Image Ad - Leaderboard
Image Ad - Leaderboard
Image Ad - Leaderboard
Image Ad - Leaderboard
Image Ad -Leaderboard
Image Ad - Leaderboard
Image Ad - Leaderboard

Image Ad - Leaderboard

Image Ad - Leaderboard