VFD estop wiring

[MrFluffy]

Thanks for the input, its driving me potty right now.
I was just reading that section (its a big manual!) and the differences between 9 and 4v trigger, with the 9v supposed to be for open ended encoders, and 4v for closed encoders. This is a open ended setup but I'm in try anything once mode now I think, so I'm off out now to nudge the dips over and see if it see's the signal coming in without the Z being connected.
I'm measuring the voltage coming out the encoder with the scope graticule, it could have a dc offset from the base also, I should go back and redo the measurement to be sure of anything about it apart from it being 1.2v from peak to trough of the squarewave.

Here's the specs I can find on the encoder in case Ive missed something stupid (I'm learning on the fly here) :-

• 1000 p/r (Single-phase 1000 pulses /R,Two phase 4 frequency doubling to 4000 pulses)
  Power source: 5-24 VDC
  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: 0-100KHz
  output: NPN type open collector output (if the collector is not connected to Pull-up resistors, no voltage output)
  Notice:AB 2phase output must not be directly connected with VCC, otherwise, will burn the output triode.
  Motor speed should be under 1000 RPM
  without Pull-up resistors
  connection:
  Green = A phase, white = B phase, red = power +, black = power -

I've been very careful to not connect the AB output to VCC, which I hope explains why its still giving a signal :-)

 

[MrFluffy]

Quadrature should be 5v or 24v, and nice and crisp.

NPN is unlikely to be suitable. You don't say what the inverter is ( or I have missed it) but most will expect 'sinking' inputs, not 'sourcing'.

A lot of inverters can be switched to V/f or sensorless vector, so they don't have to use an encoder.

Luke

Edit: just found eurotherm vector 620.

That would explain things too, sadly the vfd is mandatory encoder and cant be switched to senseless mode (unless I'm reading it wrong again).

 

[Luke]

From what I could read on the train, it does appear to require the encoder.

You could add pull up resistors and take the drive encoder inputs from just below them, but you'd still have a single sided single, without compliments.

 

[MrFluffy]

Ok, so I'd need a pull up resistor making sure the 15v encoder vcc doesnt sink more than about 5mA when its "open" because this encoder is switching a transistor on and off, so about 300 ohms. And the wiring goes vcc->resistor->A output wire from encoder -> A input wire on vfd. And the same but a second resistor etc for the B phase.
Hopefully a bit like this completely not awesome to look at diagram knocked up in a paint package.

And thats what this line in the encoder details from ebay is about :-Notice:AB 2phase output must not be directly connected with VCC, otherwise, will burn the output triode.

If thats ok, I can gnd the complements to 0V as the manual says and see if it works...

 

[MrFluffy]

I have a plan B if not as its a bit debilitating not having a working lathe I've discovered, that is steal the allen bradley vfd thats controlling the shaper that doesn't need a encoder, and fit this to that at a later date when I've sourced the correct encoder for it.

 

[Luke]

Does the manual say what the +15v output is rated for? I can't remember. Both resistors will be drawing current for 90 degrees of rotation.

 

[MrFluffy]

0-200mA total load, but that has to run the encoder circuitry also. I could measure the current draw that places and provided its less than 190mA I think that gives plenty of headroom.

 

[MrFluffy]

Ok, testing delayed because the motor worked loose on the foot as I'd only fitted it temporary to test, it has to come off to refit the gear cover etc.
But, I got to it and adjusted it up again and snugged the mounts up, and its working with the little circuit as above. It needs that encoder rotation before it will accelerate from base as unplugging the encoder from the shaft while its running drops the lathe to base rpm immediately.
I also left the dips set to 9v rather than the 4v setting, and that seems to be not affecting things unless its the cause of one of the issues below.

Well its almost working properly anyway. I've had to put the encoder count in as 500 before it spins anywhere near a plausible amount, and starting there is this weird vibration until its at >25% speed. In fact it managed to turn one of the 3 veebelts around trying to run it slower. I have a idea about the vibration being due to the temporary fudged encoder mounting bracket being a bit crap though as if I rest my hand on the bracket it will go slower before it happens.
Now at least I can take the motor off, do all the mountings and bracketry properly and fit it with at least a expectation of having a working lathe while I shake down the last of the bugs in the setup.
Thanks for helping Luke.

 

[Luke]

What does the encoder pulse train look like on the 'scope?

Do you have a tacho? What speed is the motor shaft actually doing?

I don't know this drive, but it's common to have a different understanding on what encoder pulses really are, 1000 pulses single phase can be interpreted as 2000 pulses (A and B) or 4000 edges (there being 4 edges to an A & B pulse - A rising, B rising, A falling, B falling).

Typically, the NDE of the motor shaft will be drilled and tapped for a rigid encoder coupling, with a flexible arm to give the encoder mounting some compliance.

Glad that you've made some progress.

Luke

 

[MrFluffy]

I haven't scoped the pulse out, too busy being happy the spindle span fast enough and playing with the marvel of a instant reverse button and braking parameters with it all hanging off temporary
I have a optical tacho, the sort you post a reflective dot on and it counts rpm, so I'll paint the drive pulley black and spot it today and get some actual speed readings from it.
Ive read the info on the encoder multiple times now, and it repeatedly gets itself confused on how many counts it has, its advertised as 1000pr in the specs, but goes on to mention 360, then 720 with both phases. And as above, it didnt have the Z trigger wire or capability that it had in all the screenshots also, so the figure of 1000 pr itself might be wrong also. I genuinely dont think the seller understands their product fully but thats typical of chinese sellers, they just pick up things which sell in shengzen markets and copy someone elses description and resell them from small offices, but the price reflects that...
The tacho will show what true rpm is and as long as I dont overspeed the motor for its frame all is good as the rpm display is a seperate device triggered off a seperate encoder on the lathe spindle itself after the gearbox.

For the shaft being drilled, not on this one, I was considering machining off the fan end and fitting it on the end bell with a electric fan to cool it instead as it predates encoders and that would stop the motor overheating on very low rpm use, its a original harrison 3 phase motor from the 70's ish, its a take off from a friend who converted to single phase and runs nicely though. However I'm going to machine a small stub with a large flange that locates on the dimple where the pulley doesnt sit completely flush with the end of the shaft, and put 3 radial keeper screws into the flange to attach it solidly to the drive pulley itself which is keyed to the shaft to create a stub that I can put a flexi coupling on to drive the encoder from directly, at the moment its hanging off a auxillary drive belt and whip in that belt train is also contributing to the noise which causes the jerking at low speed I think.
Ropey but it was just a quick test rig with no lathe available to make a proper encoder mount, I would have had to turn it in the mill and nothing works quite as well as a lathe for lathe work when concentricity matters. If the drive stub ever breaks off, the net result is the motor defaults to base speed not runs away uncontrollably, which is why I removed the encoder while it was turning as a test earlier.

I shall spend today freezing my fingers off in a unheated workshop at 2c playing

 

[MrFluffy]

Moving encoder from aux shaft on belt to spindle has cured the low speed vibration on start up and slow running, 1-60% now no issues though I havent rpm tested it yet so I decided that 70% felt quite fast enough for now until I can get some numbers out of it. I've also had to put the encoder count back to 1000, so maybe the belt was slipping or something odd. Just adding some buttons to the apron now for jog/start/fast stop and its ready to test in anger.
Must stop pressing reverse and marvelling at the spindle soft stop, reverse gently then soft start the other direction, repeatedly