Newly acquired Kawasaki UX150 with C-Controller

Quote from DoesItExist

I did a quick check of the wire harnessed going to JT4-5-6 and they looked mint. I don't think the issue will be on the robot side since I only moved it slightly in every direction (it's on a pallet).

I wouldn't be too quick to assume all that, that is precisely why I mentioned to carry out the test in my previous post............up to you though.

TB2 looks fine with the jumper configuration.

Whoever terminated them, removed the blue wires (X15).

If you follow them, they will go to that connector.

As mentioned, they are just the field connections:

- External emergency stop

- Safety fence circuit

These are covered in the external IO manual I posted in the C controller manuals section as mentioned previously.

Other wires from the rear of that terminal go to the internal 24V power supply unit (back left).

You may also have a led and a fuse near to the connector on the outside of the cabinet that is also connected to the power supply too.

Quote from DoesItExist

I tried swaping the motors connectors, however I didn't have enough reach on the other cables so I left JT5 unplugged and plugged its harness on JT4 or 6 (the stickers are missing)

Strange to hear that, I have done this many times without any issues of reach.

However, if that fault still remained with JT5, then it points to internal harness, umbilical's or power block.

Quote from DoesItExist

I tested all the wires on the controller plug and harness plug with a DVM and none of them were shorted to ground.

Each harness on the arm is separate and the internal conductors are shrouded with an internal shield.

Using a DVM may not yield any issues due to the low voltage and impedance existing with DVM's.

Not to say a DVM is not a capable tool to cold check with though.

If you want to be 100% with harnesses (especially old ones), the only good method is to use a megger on them by disconnecting them on both ends and shove some HV across the conductors to check for cross conductor leakage.

I have seen many an old robot with internal harness shorting to the internal shield when in certain positions where the harness is compressed/under tension and an internal conductor is damaged.

With you moving the arm successfully prior to this, it is very conceivable that an internal harness to JT5 could have been stretched/compressed causing damage.

Quote from DoesItExist

I think this might be an issue on a board or current overload sensor since the motors are not even powered up... I'm going back to the troubleshooting manual .

However, it's not for me to determine whether your cables are good or bad......only you can check on them and decide whether they are good or not.

Check the connections to the power block are secured (all Kawasaki have clicked connectors) so if you pull on them, they should not come out.

Then all that leaves is the power block itself, or the AVR supplies may be low - so it's worthwhile checking those using a DVM on the CPU board test pins and make sure they are in spec.

If not, you can use the potentiometers to adjust them so that they are within range.

The AVR supplies and test pins on the CPU board are referenced in the troubleshooting manual.

Certainly sounds like an intermittent cable issue, which is good news really in comparison to potential hardware failure.

Depending on how the 2nd user umbilicals were used/mounted in the field, areas around the connector gland could have internal damage if they were not located or improper strain reliefs were not used.

Looks like your making progress though.......

Visual inspection on umbilicals will not suffice as a categoric check.

To 100% test them, you would be better removing them completely and independently megger each conductor (takes time, but you never know with 2nd user kit).

I wouldn't be so quick to rule them out - X5/X5A

I would check the internal housings for foreign material too, you could be really unlucky and have something wedged in there.

I would also check inside the cabinet, where the umbilicals connect to, it is possible a wire or foreign object is in there (never come across that, but possible).

I would also check the connector to the power block XM3SA to make sure that is connected securely and check for shorts there too.

Quote from DoesItExist

I noticed the damaged connector has a bit of play but still don't think this is the issue.

What's this?

If it is an umbilical to the controller, then this needs to be secure with the clasps.

Are you saying this is damaged?

Can you tie wrap it all tight if so and retest?

We're still not in a position to categorically state where this issue lies with the in frequency of just when this error is appearing.

She's being very stubborn, come on lady, flicker those eye lashes for us....

Can you check the 1GB board (under the metal cover mounted to the power block).

Check and confirm all cables are securely connected.

Specifically CN10 as this is directly connected to the current sensing modules for JT4, JT5 and JT6.

Can you also measure the 1GA voltage test pins and report back the actual DC values from them?

My spider senses are still leaning towards the umbilicals or inside the connectors of the umbilicals though.....

Yes the connector damage still looks to have enough purchase on it via the clasping lugs.

The voltages look within spec although the 1GB voltages look a little low to me.

Maybe worth putting some test clips on the 1GB pins and adjust the pots on the front of the AVR for -12 and +5

But they will only need the smallest tweak and I emphasise this not even a segment.

Quote from DoesItExist

I now have the X5 Umbilical unplugged but the JT5 error keeps coming back.

Yep, you mentioned this and this is where she is throwing in a saucepan................as it is logically pointing to the hardware in the controller as opposed to the umbilicals based on that.

The attached image red circled is CN9 and CN10.

- CN9 is the current sensor feedback from JT1, JT2, JT3 to 1GB.

- CN10 is the current sensor feedback from JT4, JT5, JT6 to 1GB.

Yellow circled is CN1 and it does not appear connected fully - see the open wing, snap that baby in.

As your error is pointing to JT5, then CN10 connection is definitely worth checking by removing it, cleaning it all up and reconnecting it.

Those connectors are not easy to remove - ie impossible to vibrate loose due to a lever on the side.

You will notice a small white lever on the right side of them, which retains them in place.

So be careful when removing it.

I would get some IPA or switch cleaner to clean the pins and plug to remove any crap and reconnect.

See if that helps any.

Other than that, it looks like there is some problem via the current sensor on JT5 in the power block.

The current sensors are hall effect (blue doughnuts where the output phases pass through).

It may be that the sensor itself has a dry joint soldering it to the board.

This is where it's pointing more towards now.

Quote from DoesItExist

I noticed the TP had blacked out again and restarted the controller.

Never heard of that before.

You are aware the TP backlight shuts off after 5 or 10 mins (I can't remember precisely) of inactivity.

When you total all these symptoms up, it's as though there is some power supply issues to the logic side of the electronics.

Can you take some voltage readings again tomorrow and also check the incoming supply phases to earth and confirm they are all balanced and also phase to phase to readings.

I would consider tweaking those AVR voltages too, the 1GB values appear quite borderline to specifications.

It may just be something like this that's making her not wanting to play properly or conceivably be the AVR starting to fail.

Pity you don't have access to some spare parts.

I do know a company here in the UK that carries lots of 2nd user parts (as they buy ex jaguar controllers) and I can give you a name and contact email if you require - but they are likely to be expensive due to the obsolete nature of the parts.

............................... .........................

My word she's really throwing her legs out of bed isn't she.

Ok, so let's sum things up........

1. Initially you have this to salvage parts from.

2. Plugged stuff in, and discovered it moved.

3. At this point, no real exercise tests/warm up tests or a full operational check was carried out.

4. Errors may have been there, but no full testing was done, cannot rule this out from the start.

5. Then an issue with the TP screen - which appears to have been resolved with fuse bridge.

6. Since then, regular JT5 Amp overcurrent errors intermittently.

7. Small AVR tweaks made which are good IMHO.

8. Now you are getting JT5 motor overload error, JT5 position envelope error and suspicious noises.

OK, looking at the errors:

JT5 amp overcurrent:

- Current sensor from power block to 1GB, 1GB reports too much current on JT5.

- Possible motor stator shorting to ground.

- Possible power block failure.

- Possible harness failure.

- You say this error can appear on power up of the controller, before you turn on motor power.

- You say this error can appear even though the umbilicals are disconnected.

JT5 motor overload:

- Force acting on the joint preventing motion.

- Possible brake not releasing or mechanical link issue (gear box, motor spline, drive rods).

JT5 position envelope:

- commanded position is different from actual position - possible brake not releasing.

- Possible brake not releasing or mechanical link issue (gear box, motor spline, drive rods).

Amp overcurrent, motor overload, position envelope error is usually related to the arm:

- Motor brake failure.

- Motor failure where stator shorted to ground.

- Internal harness failure.

Motor overload and/or position envelope will never appear unless the robot is attempting to move.

Throwing the suspicious noises into the mix, it maybe the brake was not releasing and the noise was the motor shaft rubbing against the brake plate.

- Check MCB F5 inside the cabinet, cycle it a couple of times/blow some air/switch cleaner.

- Then try and get a buddy, and test each joint using brake release buttons for each joint.

- JT4, JT5 and JT6 may be difficult to move due to back driving through the gearing.

- JT1 is easy enough.

- JT2 and JT3 will drop due to gravity, so suspend the arm FIRST if you are going to test JT2 and JT3

I am hoping the suspicious noise, motor overload and position envelope are the result of the brake not releasing due to F5 (this was a common aging issue on these controllers - MCB becomes crusty).

Next time you power it up and without any errors, instead of trying JT5, can you try the other joints instead?

Do you know what I mean when I say, select Joint Interpolation when trying to move it manually?

Quote from DoesItExist

I'm not seeing MCB F5 nor finding it in the troubleshooting manual. Can you help me locate it ?

There are 3 MCB's on the inside left of the controller (resettable breakers) near the big contactors.

F3, F4 and F5.

F5 is for the brake supply.

Just a question to, what voltage are you supplying to the controller?