[joeavaerage]

Hi,
I think before you waste all your time making a control box you should set up just one servo and run it so you can 100% how to control them.

Craig

[dazp1976]

Hi,

But the encoder plugs into the servo drive, not the BoB. The encoder of the servo is surely matched to the servo drive.

Craig

And indeed it is and they are. But..

I think we may have our wires crossed.
The servo motor itself is plugged in to the UVW,PE connection and the encoder is plugged into the drive next to that. into CN4 encoder port. That's all fine and I've spun it up using the panel, runs really smooth.

It's the pulse feedback output signals that go to the bob via the DB44 control signal cable, that's plugged into the CN3 port that I'm talking about.
CN3 is the one with all the IO controls.

The A+B+Z+ , A-B-Z- , CZ & GND

Pg 14 section 4.4.5
Pg's 9-10 section 4.3.2
pin no's 9 through to 15 & no24
http://servo.xlichuan.com/Private/Pr...6713131150.pdf

Related configuration parameters:
PA_044:feedback pulse doubling molecule
PA_045:feedback pulse division octave denominator
PA_046:Feedback pulse logic inversion
I'm going to read up regarding the perameters.


These Lichuan things seem a bit more refined and have a lot more IO compared to those 'other' Chinese drives with only a DB25 output.

[joeavaerage]

Hi,

It's the pulse feedback output signals that go to the bob via the DB44 control signal cable, that's plugged into the CN3 port that I'm talking about.
CN3 is the one with all the IO controls.

And why do you need those?

Your UC300/BoB is not a feedback controller so it does not need encoder feed back from the drive.

Craig

[dazp1976]

Hi,



And why do you need those?

Your UC300/BoB is not a feedback controller so it does not need encoder feed back from the drive.

Craig

That'll do me then. I could see it only being needed for rigid tapping anyway. That's not for me that.

Testing it under Mach control tomorrow. Have a cheapo 5 axis bob and pc spare.
Then I'll have a little fiddle with the 300eth if above test successful.
Dual boot- xp for mach over parallel. Win7 for uccnc.& mach over ethernet.

Relays and wires everywhere on the test bench (lounge floor), looks a mess lol.
Been sat there like that for months because I couldn't be bothered.

[joeavaerage]

Hi,

I could see it only being needed for rigid tapping anyway.

You don't need it for rigid tapping either, you can use an AC servo as spindle and use it as C-axis, in which case you don't need encoder feedback to the controller.

With a C axis it is coordinated with the Z axis so it does a helical interpolation, aka rigid tapping.

Craig

[dazp1976]

Hi,
You don't need it for rigid tapping either, you can use an AC servo as spindle and use it as C-axis, in which case you don't need encoder feedback to the controller.

With a C axis it is coordinated with the Z axis so it does a helical interpolation, aka rigid tapping.
Craig

Thanks for the info.

Daz.

[joeavaerage]

Hi,
this is the whole point of modern AC servos.......is that they should be easy to use.

Years ago a servo had an encoder that feed back to the controller and the controller had to provide a PID function with all the complexity and tuning details that entails.
Nowadays the servo encoder goes to the servo drive and the servo drive does the feedback PID loop, with vastly better tuning facilities than was ever available hitherto
and the controller need only be open an loop step/direction type.

So simple they have become that you can treat them like an open loop stepper, just signal to the drive a certain number of steps in a given direction THEN allow the
servo and drive to get on with the business of getting there. The only time you'll hear from the drive is if it fails or encounters an overcurrent /overvoltage fault/ following error
fault, and with any sort of luck and with some decent settings being applied that will be almost never.

The only complication that comes from using them is signalling them. You commonly require differential signalling for the step/direction commands in order to take advantage
of the high speed and resolution servos have. In addition you will have a few more inputs, things like servo enable, servo reset that you may not have encountered with open loop
steppers. Also there are more outputs, typically at least one fault or alarm output, but its possible to have multiple alarm outputs.....so you can treat different alarm conditions
differently. Its possible to monitor other servo conditions that can be attached to a digital output that would allow the controller to monitor the servo condition, Zero Error Window is the
most common use but there are others.

Once you become accustomed to the somewhat wider and detailed signalling schemes used by servos you will realize just how easy they are to use and how flexibly they
can be controlled and used you will ditch all steppers thereafter.

Craig

[dazp1976]

And indeed it is and they are. But..

It's the pulse feedback output signals that go to the bob via the DB44 control signal cable, that's plugged into the CN3 port that I'm talking about.
CN3 is the one with all the IO controls.
The A+B+Z+ , A-B-Z- , CZ & GND
Pg 14 section 4.4.5
Pg's 9-10 section 4.3.2
pin no's 9 through to 15 & no24
http://servo.xlichuan.com/Private/Pr...6713131150.pdf

Related configuration parameters:
PA_044:feedback pulse doubling molecule
PA_045:feedback pulse division octave denominator
PA_046:Feedback pulse logic inversion
I'm going to read up regarding the perameters.

Had no time to get onto anything lately.
Only thing I've done is unpacked a ucbb board to have a look.
It has a few high speed optos and I noticed some 74HC14D chips on it.
Upon research these can be used as line receivers.
So, these ABZ CZ GND connections can go straight into the board if I needed.

Craig is right about making equipment pretty straight forward these days.

The issue is only having the confidence in yourself!!

[dazp1976]

Gawd dam it.
What am I missing?.
Most of it's working fine, The DI side, Servo on relay switch, reset, direction changes CW/CCW, Estop.
Analog 0-10v the rpm gain is now spot on. Mach runs it all perfectly.

When it comes to the D0 outputs, nothing seems to trigger regardless. Simulated a couple of faults like the main power disconnect and encoder unplugged but it changes nothing.
Any deas?

All wired exactly as per the diagram. Relays are all module types with opto's and high/low jumpers.
Output relays are set as high triggers as the D0 (supposedly) emits +ve voltage.

[joeavaerage]

Hi,
if your servos are anything like my Delta servos then a DO must be 'logically connected' to an alarm.
That is there are eight parameters in the drive that attach each of the eight DO's to a condition.

Craig

[joeavaerage]

Hi,
page 18 of the manual lists the parameters for the DO's:
PA_088
PA_089
PA_08A
PA_08B
PA_08C
PA_08D

Page 19 describes the function of each of the range of 17 possible settings you can make. Frankly the range of alarm conditions is poor as is the description of
what the offered settings mean.

Craig

[dazp1976]

Hi,
page 18 of the manual lists the parameters for the DO's:
PA_088
PA_089
PA_08A
PA_08B
PA_08C
PA_08D
Page 19 describes the function of each of the range of 17 possible settings you can make. Frankly the range of alarm conditions is poor as is the description of
what the offered settings mean.
Craig

Tell me about it. If you can excuse the French, it is a little piss poor.
It's in speed mode. The Inputs are all working perfectly, have the analog nearly bang on in Mach within 10rpm's.
Didn't need the 'intspd' inputs. PA_006 set to 2 is direction control (DI5 atm). Relay off =CCW, relay on =CW. (M3/M4).

The outputs however are as though there isn't a positive connection there for the opto's to output to relay trigger.
They're low level output (com-), high level trigger (24v+)
As it stands I have in order:
Servo ready
Servo Alarm
Spd arrival
Brake release
Zero speed
Torque limit
I only need 2 or 3 of them tbf.

Servo ready should do at least something with the main power disconnected. I took the main out and left control power in. Fired it all up and still no change.

The only thing I've stumbled on so far (fishing now) is PA_08E. (bottom of page 40)
IO polarity reverse setting. Available= -32/68 to 32/6/. As it is it's at 0. Says something about the higher 8 bits correspond to D0, Lower 8 bits correspond to DI.
Can't sse how a change could hurt. The inputs are completely isolated. (well, apart from the com- connection).

[dazp1976]

I'll prob just have to go with it in the end. Sit and babysit the damn thing in case it needs an estop triggered by hand Ugh!.

[joeavaerage]

Hi,
I was of the opinion the DO's were opto-transistor outputs, that is to say they are neither high or low until or unless you apply power,
so are rather either conducting or non-conducting.

Craig

[dazp1976]

Hi,
I was of the opinion the DO's were opto-transistor outputs, that is to say they are neither high or low until or unless you apply power,
so are rather either conducting or non-conducting.

Craig

That's what I'm thinking. Don't know much about transistors at all.
I was actually leaning towards them being wired like this internally in the dia below.
Using CMOS gates but they're not getting +ve power for some reason.
https://www.cnczone.com/forums/attac...d=464362&stc=1
I'm definately fishing now. Out of my depth. It was going so well too. (nuts)
If I apply my own power to it wouldn't I just set off the relay switch permanently?.

[joeavaerage]

Hi,
you are way overcomplicating things. That circuit you've posted is an input circuit at the left, though an opto-coupler to the 'internal' circuits, in short an
input circuit. You don't need or want an input circuit at all.

Section 4.4.3 page 13 shows the situation well.

Firstly note that DO1,DO2,DO3 and DO4 have one circuit which differs slightly from DO5.

What they all have in common is hat they are ALL photo-transistor output. The photo-diode is hooked up to the internals of the drive, according to the parameters you set.

Lets say for instance you set parameters such that DO1 is logically connected overload detect circuit. Under normal operating condition, ie no overload, the photodiode is inactive
and therefore the photo-transistor is non-conducting. If an overload occurs however then the photo-diode is active and therefore the photo-transistor conducts and the collector 'pulls low'
sinking current to COM. You could use that sinking current to turn on a light, or a relay, or pull an EStop line low.

DO5 differs in that the collector AND the emitter of the photo-transistor are bought out to separate pins. This is rather akin to having an isolated set of relay contacts.
Unlike relay contacts you can only pass current one way in a transistor and you'd be a fool to try to pass more than 20mA, for fear of blowing the transistor inside.

As I posted earlier the manufacturers deliberately make it easy to interface to their drives, you don't need a photo coupler....they're already built in.
Just use the outputs as: non-conducting= no alarm condition conducting=alarm condition

If you want to 'make it hard for yourself' I would suggest you stick to the age-old way that your mother warned you about....leave optoisolators for those who've
already gone blind.

Craig

[dazp1976]

Hi,
If you want to 'make it hard for yourself' I would suggest you stick to the age-old way that your mother warned you about....leave optoisolators for those who've
already gone blind.
Craig

Lol (wedge)

Bear with. You telling me 'sink to gnd' got me reading last night.
Ultimately I think my main issue here is using multi module relay, opto units. Now why is that, ahhh......

So take a simple opto for example. Normal working is P1=vcc, P2=gnd, P3&4 are load, add power and the load comes on.
Other diagrams show me that for MY outputs I need to put vcc into P2, go out to the servo from P1, then through servo opto into gnd.
Actually makes sense as this basically puts all the diodes in a simple parallel connection. Yes?.
So I'm guessing that means they're in NC config. When servo errors it goes NO and sets off the trigger.

With the multi module units the vcc's are basically fixed to P1 and signal from servo etc goes in P2. The diode is now BACKWARDS!!!!
BOSH!! There you have it!.
I can either figure out how to wire them in an opposing way or bin them off and get a simple thing that I can wire as I see fit such as this below.
Or get some individual ones.
I have the types with the leds etc all over them which all vcc into P1.
https://www.cnczone.com/forums/attac...d=464382&stc=1
Mine is:
https://www.cnczone.com/forums/attac...d=464386&stc=1

But wait a minute. I have recently got UCBB breakout boards. Well here's the pnp diagram:

https://www.cnczone.com/forums/attac...d=464384&stc=1

All the inputs are like this with both +/- ports. This is going vcc24v+, into opto P2, out from P1, into switch (servo output xx), out to gnd.
I can just wire the stupid thing straight in can't I.
YES?.

If there was an emoji banging his head against the wall... I'd use it.

Darren.

[dazp1976]

Actually makes sense as this basically puts all the diodes in a simple parallel connection. Yes?.
Darren.

I meant in 'series' not parallel in the comment earlier
Couldn't edit the post.

[dazp1976]

Hi,
I was of the opinion the DO's were opto-transistor outputs, that is to say they are neither high or low until or unless you apply power,
so are rather either conducting or non-conducting.

Craig

Hi Craig.
You can shake your head iat this one if you wish!
Solved!!!

Spent hours last night wiring everything into the UCBB board.
Put the servo outputs directly into the boards inputs.
Fired up the servo and voila. 'servo ready' and 'zero spd' opto's lit up.
So when it's ready it's NC connection, when it's not it's NO.
This is actually perfect for wiring directly into estop.

I did a little test by pulling the main power out and leaving control power in.
Opto's did not light up indicating a problem and servo was 'not ready'. Perfecto.

Here was my problem all along:
I got together all of the various modules I'd been trying out and my probes.
I had a good look over the 8 chnl relay module I initially had the outputs connected to.
Had a eureka! moment.
Right. So I have the +/- vcc's into module power input. Servo ready connected to trigger 1.
If I follow the circuit backwards it goes GND- through the servo output- relay trigger- vcc+ of module to complete the circuit.

That's actually a low/gnd trigger...............
I had the jumper on my relay module set to HIGH. Thus resulting in an incomplete circuit.
That's all it was!!!!!! omg!

Thank you for all your help on this Craig. It was you mentioning 'sinking to ground' that put me onto all this.
Also mentioning the DMM servo's. The wiring diagram on that was more explanatory.

I now have a fully functional system and had 0 accidents due to your help.
You're right about equipment being made somwhat easy to hook up these days (now that I've figured out where I went wrong).

Have to transplant everything into a new control box next. Ugh!

[joeavaerage]

Hi,

That's actually a low/gnd trigger...............

It's called active low and is very common in industrial control hence servo manufacturers tend to use it.

Craig