Gem e2 immobile - error 45, error 81 in high mode

GEM Forum
1

I am trying to fix a 2014 Gem e2 that spontaneously stopped working. Motor and controller details below.

The user at the time said that the buggy was reversed out but when the mode was changed to go forward and the accelerator pedal pressed, there was no motion and error 45 was displayed.

I’ve taken a look and the motor doesn’t run in any mode (high, low or reverse). With the front wheels on axle stands, handbrake off, the master disconnect switch on and the ignition switched on, I hear a click from the master relay in all modes when I depress the accelerator pedal, but the motor doesn’t run. In “high” mode, a very slow application of the pedal causes error 81 to be displayed. If the application of the accelerator pedal is quicker, error code 45 is displayed. In “low” or “reverse” modes, no error code is displayed.

I interpret Error 45 as reporting “Armature transistor did not turn off properly: Replace the motor controller”. I don’t know where to start with this.
I interpret Error 81 as reporting “No tachometer signal is detected: Defective wiring between tachometer and the motor controller. Replace tachometer.”. I checked the wiring between the tachometer and the motor controller with my meter and its fine. I measured 12V across the black and red wires. I tried operating the buggy with the tachometer disconnected and the behaviour seemed the same so its possible that the tachometer is broken but I can’t see why that would prevent any motion at all in the motor or cause error 45. But I suppose its possible.

My strategy has been that the error 45 is likely to be the real culprit but maybe investigating the error 81 (which might be easier) might point me to a common fault. I was thinking perhaps of a dodgy connector or damaged wiring. But I didn’t find any sign of damaged wiring. I checked that the connectors were securely seated.

Motor: Part No.GU3-4002, 2412501 class H
Controller: FSIP catalogue number IC3645SR7A353T6

One rather surprising (but probably unrelated) discovery is that the big orange cables between motor and the controller appear to be swapped. The one that is labelled “<- MA1 CA1 ->” actually goes from Motor A2 to Controller A1. I don’t believe this is relevant to the problem I am seeing - its inconceivable that someone came along and swapped the wires.

Can anyone suggest a route forward? After seeing videos of the inside of the controller and the effort required to restore suitable thermal sinking for the MOSFETs, I am reluctant to dismantle the controller if it can be avoided.

Thanks,
Richard

2

But I suppose its possible.

Is there anyone out there with a similar vehicle who can confirm that their vehicle is not immobilised by disconnecting the speed sensor (tachometer)?

If their vehicle still runs, I can just disconnect the tacho and ignore the 81 error (though I’m not sure where that leaves me).
If their vehicle doesn’t run then I can focus on the tacho as a likely source of the problem.

3

The early Gen1 cars (00-04) will sometimes run without a speed sensor. They had the T1 and T2 controllers.

Your 2014 car has a T5 controller and a bad speed sensor will fault with a -81. The car will not run without it. The fun police worked their way too far into the programming to allow the car to run by simply unplugging it. Sometimes it will run for a few feet/yards/blocks before it sees it is not getting signal from the sensor then throw the code. A good sensor will generally register 1~2 mph on the Dash Display by pushing the car a few feet forward or backward right there in the garage.

If you checked the sensor plug down at the motor then also check the 23 pin connector up at the controller. Otherwise- replace the sensor. Checking wiring at the 23p controller plug is tricky because it involves probing the controller pins. This action is suspect as the meter probes tend to stress the internal pin wipers and compound the problem. It is ironic that the very action suggested in the factory service manual causes more problems. (we can dive into that later)

The real concern I see here is the code 45 you are being presented with. This is usually due to a weak or failing transistor in the controller and a service is required.

The interesting thing about this code is that sometimes it can be triggered by a stuck brush in the motor. Try removing one of the A wires off the controller and see if the fault still appears. No, your car will not run- This is just a test. Neglected motors always benefit a good blow out with compressed air (so many ways I could have gone with that).

Your observation of the A1/A2 wires being swapped tells me that someone else has been messing with this car. It is probably not the cause of the fault, but this swap alone would make your car run backwards unless the F1/F2 wires were also swapped.

What is the history on this car?
Is this a new car to you?
Are you just servicing this car?
When was the last time the motor was serviced?
What battery is in this car?

Code 45 ia also seen by the hot rod guys. Higher voltages and hotter programming numbers push limits of the controller and the transistors sometimes get cranky.

Report back with what you find.

1 Like
4

Many thanks for your response. I’m encouraged to hear that a faulty speed sensor could be sufficient to immobilise the vehicle.
Is there any value in temporarily connecting the sensor input on the controller to +12V (or GND?) to see if that can fool the controller into thinking that there is tachometer present? If the vehicle then runs, I can be confident that the problem is with the sensor.

A good sensor will generally register 1~2 mph on the Dash Display

Since the vehicle is up on axle stands I can try spinning the wheels. If I get a reading on the speedometer display would that suggest the tacho sensor is ok. Is the converse true? I should check.
But I didn’t see error 81 on the display in either Low or Reverse modes, which surprises me.

If you checked the sensor plug down at the motor then also check the 23 pin connector up at the controller

Yes, I popped 23-pin connector out and went to the pins on the back of it. I measured continuity from those pins to the three female connections on the sensor connector by the motor.

The real concern I see here is the code 45 you are being presented with.

Indeed. But baffling (to me) that it only occurs in High mode, not Low or Reverse. I am trying to think if that is significant. Could it be that it doesn’t get triggered until a certain amount of power (more than would normally be available in the other two modes) is demanded?
I am playing with an idea where the immobility issue is due to the speed sensor. Fixing that would allow the vehicle to be used in Low and Reverse mode (since error 45 doesn’t occur in these modes). Using High mode might need to wait until the transistor issue could be addressed.

sometimes it can be triggered by a stuck brush in the motor. Try removing one of the A wires off the controller and see if the fault still appears.

Excellent suggestion. I’ll give that a go next time I am with the vehicle and report back.
I do know that he motor is not seized or stuck. When I rotate the wheels I see the motor armature rotating.

What is the history on this car?
Is this a new car to you?
What battery is in this car?

It is owned by an Outdoor Education Charity where I volunteer. I believe that it was gifted to the charity by a university at the completion of their research into self guided vehicles. It seems possible that the previous owners may have messed with the wiring. I don’t know what other modifications they may have made.
After being used around the site for a while, the vehicle stood unused for many months at the end of which the batteries were dead. These were replaced. I will need to visit the vehicle to find out what the batteries are. I’ll get back to you on this.

Are you just servicing this car?

Haha. My usual activities as a volunteer involve hedge-laying, strimming, changing bedding, etc and all manner of menial tasks. It was found out that I also volunteer at a “Repair Cafe” and have a history of fixing stuff and was asked to try to get the vehicle going when it failed to work after the batteries had been replaced (the major issue was a missing fuse which meant that the handbrake release sensor was disabled). That was a couple of months ago. Then they had a problem with charging and I was asked to sort that. Now its not working at all and they have asked me again to look at it. It is said that no good deed goes unpunished. Hahaha.

When was the last time the motor was serviced?

Hmmm. I can try to check but I may not get an answer.
We are in the UK and I am told that it is difficult to get technical support. Hence why I am looking at it.

Stuff for me to do:

  • check batteries
  • spin wheels and observe speedometer reading (might need a colleague for that…or a mirror)
  • remove A wires from controller and see if error 45 still occurs
  • ask about servicing

Many thanks again for this valuable information.
Richard

5

Other potentially significant information

  • The motor doesn’t even twitch. I can envisage a situation where -81 would be generated if the controller was expecting to see pulses from the speed sensor and didn’t. But in this case it seems to be able to to detect a problem without the motor ever moving.
  • I bravely put my voltmeter across the A terminals on the motor and slowly depressed the accelerator pedal. I saw no voltage of any significance. There was a very small disturbance (maybe 0.2V) when the relay clicked but nothing else. At this point neither error had been displayed.
6

Don’t bother with asking about motor service history, We can assume “Never” (or alt least not in a while). If you said “a couple months ago” this would be different. Depending on which motor is in this car (post pic?) you may have a band clamp on the brush end. Take that cover off and inspect brushes. It is easier to do if you remove the motor, but more effort.

These motor hate sitting. Especially if near the coast.

If still up on stands, turning one wheel will generally cause the other wheel to spin the opposite direction and not the gearbox/motor which is what is needed). This is due to the nature of the differential in the gearbox. Have an assistant hold that wheel stationary and turn only one wheel to check speed sensor response on the dash. For lack of available volunteers you can block under it to hold it.

7

post pic?

I have pictures…

image
image1040×585 90.6 KB

image
image1040×585 43.7 KB

image
image1040×585 59.9 KB

image
image1040×585 93.3 KB

I’ll get a picture of the batteries next time I am with the vehicle.
We are in Oxfordshire which is a good 100 miles from the coast in every direction.

8

Wow- a T6 Controller?!?!? I will need to see if I can find a page on that one.

AMD motors are great, but also the ones that have the tightest tolerances on their brushes and the biggest sticky brush violators. They are also one of the toughest motors to fully disassemble without some special tools. Getting in to service the brushes is fairly easy tho.

9

battery

image
image1040×585 67.5 KB

10

i’m with the vehicle now.

if i spin the wheel with my foot with power on and lean back to watch the display, i can get up to 2mph displayed. this suggests to me that the speed sensor is working…
…which ruins my best theory

i also removed the A1 connection on the controller. this seems to get rid of the -45 error, but other than that the behaviour is the same, i.e.

  1. vehicle does not run in any mode
  2. click heard from relay in all modes when pedal pressed
  3. gentle pressure on pedal in high mode produces -81 error
  4. no errors in other modes except if -81 has already been triggered in high mode. therafter low mode will produce -81 error until power cycled. reverse still produces no error

this behaviour is unchanged if i disconnect the speed sensor

i have pictures of the brushes which i will post later

any thoughts on what to look at next?

11

Your initial post mentioned

I’ve taken a look at the motor".

What does this mean? How deep did this go?

I have pictures of the brushes which i will post later

Not just pictures, but inspections should include lifting up on the brush spring and making sure each brush slides nicely in the holder. (Easiest to do when motor is out of the car.)

With a good light, look at the armature.

Depending on what you have for a meter, give it a resistance check. All tests with motor disconnected from control.

Motor test

  1. A1 TO A2 should show less than a few ohms (.1 to .2) Almost a dead short.
  2. F1 TO F2 should show a few ohms (1.3~1.5)
  3. ALL THE ABOVE TO CASE should show nothing (OL)
  4. F’s TO A’s should show nothing. (OL)
  5. ANY DEAD SHORTS motor is toast. (000)

If you have any spare parts, It might be worth swapping in a known good controller (T5?) and/or a motor to see if problem changes.

12

Thanks for this.

I’ve taken a look at the motor".

it was “I’ve taken a look [at the buggy] and the motor doesn’t run in any mode”. Sorry if that wasn’t clear.

I’ve now been back and looked at the motor…by which I mean that I removed the guard and taken pictures of the apertures

image
image585×1040 44.5 KB

image
image585×1040 54.9 KB

image
image585×1040 88.6 KB

image
image585×1040 89.8 KB

inspections should include lifting up on the brush spring and making sure each brush slides nicely in the holder

I will need to go back and do this.

Motor test

I will take my Fluke (and some chunky test leads) with me and check these values.

Unfortunately, I don’t have access to replacement parts.

…which ruins my best theory

I had been hopeful that a faulty speed sensor was causing the controller to prevent movement. That doesn’t look to be the case, and now I suspect that the -81 is caused by the controller thinking “I should have seen some pulses by now but haven’t, so I’ll raise an error.”. Does this imply that the controller thinks that the motor should be moving?

The absence of the -45 error when I disconnect the A1 connection on the controller would seem to indicate that the connection between the controller and the motor is a factor. That suggests to me that the controller is trying to drive current through the motor and something isn’t quite right.

I’ll get those resistance measurements for you.

Thanks for your comments

13

Wow- This motor sure looks like it could use some attention. What is the history of this car? Does it spend time at the beach? Even if it is not the problem, it would benefit to have a good clean out and comms cleaned.

Also- this motor should have a temp sensor wire (2 pin harness/plug) down next to the speed sensor. Unplug that and see if your -81 clears. It should not be related to the -81, but stranger things have happened. You will probably get an over temp warning on the dash, but that won’t keep your car from running.

14

I don’t believe that the vehicle has ever been near the beach. I believe it was a used as part of a research project into self-guided vehicles by the University of Oxford and was gifted to the Outdoor Education charity who now own it.

I recall seeing that temperature sensor and that when I disconnected it, I saw a temperature alert on the display. My recollection is that it didn’t help but I will recheck.

In Low mode (and Reverse), I see no errors on the display yet the vehicle doesn’t move.

15
  1. If you have access to compressor, blow the carbon dust out of the motor.
  2. Check each of those brushes and make sure they are sliding.
  3. Get a can of electric motor cleaner. See if you can clean the comms. tow/push the car around if you have to. The cleaner and brushes rubbing may clean off the comms.
    Better → If the motor is out of the car, see if you can pull a set of brushes back, slide some 120 grit emery cloth in/under them, drop the brushes back down and turn armature for several revolutions.

** Note, this is not a substitute for a real motor service, but it may help with diagnostic purposes.

That temp sensor was just a hail mary play. Sometimes the insulation breaks down and the sensor shorts to the case and does weird things. Bypass the error by connecting the two wires in the plug and the warning will go away.

16

If you have access to compressor, blow the carbon dust out of the motor.

I will ask.

Check each of those brushes and make sure they are sliding.

The motor rotates fine, but I will make sure I check that the brushes slide ok. Is it just a matter of pulling up the springs and giving the brushes a jiggle?

Get a can of electric motor cleaner.

Would this do the job?

image
image585×1040 86.7 KB

Stuff for me to do

  1. Measure resistances
  2. Blow out motor
  3. Check that brushes slide

Thanks

17

Just wondering. Would spraying liquid cleaner into the motor risk flowing carbon dust (now liquid crud) between copper bars and other places it shouldn’t be and cause a short? Or maybe a lack of resistance?
Now I’m not an electrician or anything. Just wondering.

18

I plan to go and measure the resistances of the motor today.

Also…I have a 200W bench supply. Is there a way that I can use that to test the motor? I am thinking that I could remove the A1 and A2 cables and then switch on the vehicle. [I am hopeful that the field windings would then be powered.] I can then use the bench supply connected across the A1 and A2 on the motor to gradually ramp up the voltage to the motor windings. [The vehicle is electrically floating so I shouldn’t have any problem with earth.]
How much voltage/current might I have to apply to get some movement in the motor?

I also have a 12V lithium battery from a golf trolley, but iI worry that it might explode up if I put that across the motor connections.

19

Possibly, if we were talking about a component that is affected by trace electrical signals such as a circuit board where the difference of a few ohms or millivolts will change the outcome of a pass/fail decision, but if I think electrons would take the path of a much latger copper coil with less resistance.

But also note in my step 1 (blow the carbon dust out of the motor) was to blow as much of the loose stuff out first.

Also- hopefully a follow-up wash of the motor cleaner(specifically) will not wash off the varnish of the motor windings but more to get some of that crap off the comms. Other cleaner sprays like Brake Cleaner might do a better job but the solvents might also remove the varnish and ruin the motor.

I’m not so worried about the thin layer that would result when there is far more of it packed in the gap between the motor comms (piano keys) from naturally running these things. I’ve often wondered why those turn into a direct short all the way around. On the motors I’ve cleaned out I ran a thin blade down in between each segment and carved all of that out. Made me feel better.

20

One step at a time. I like to peel back the sheets one at a time looking for the cockroach rather than wadding it all up and taking it outside. That way I know I got it. (Now that I think about it…this outcome of this analogy has the outcome work either way possibility of possibly working out, but also that the little bastard might still get away)

I think your 200w supply might not be enough but it might get you some action. I’m not sure what it will tell us. You would need an assistant as Field windings would not get energized until the pedal is pressed. You could use a battery and direct connect to the A1/A2 but also need to jump to the F1/F2. Probably run the Field in series.