Acting like batteries are not charging.....sometimes

[cptaylor]

96 TXT. New batteries, new 4 AWG cables, new F/R switch.

Golfcart drives great.....sometimes! other times it acts like the batteries are completely dead. This one has me stumped. I can hook up the charger and let it go until it shuts off. Half the time the cart will go like no other. The other half, I may be lucky to get it to go 20 feet before acting like it's dead. I'm watching the batteries with a Fluke 83 DMM (calibrated) and can't see enough difference between when it's working good and when it's not to see anything that would raise a red flag. Battery voltage does not fall below 37 Volts.
When I first plug it in, the current meter will jump up to say 15 to 20 amps and go down to a few amps. It takes several hours to charge. Water level in batteries is good. The batteries were broke in correctly. No one battery seems to be weaker than the other.
What other points should I check with DMM and what should I expect? I am particularly interested about the controller and what to see at those points. This is not a programmable controller.

Any suggestions appreciated....tired of having to push it back up hill.

[JohnnieB]

Sounds like your batteries and charger are okay.
Could be the F/R switch (New doesn't mean good or properly adjusted)
Could be the controller.
Could be the ITS.

Connect your DVM between the M- and B+ terminals on controller and monitor voltage.
Should go between zero when solenoid first clicks and full pack voltage when pedal is on floor.
See if there is a difference when cart is running right and when it is misbehaving.

Here is a schematic with the high current circuit highlighted.

[cptaylor]

ok, got it when it was and wasn't acting right.
BIG difference in voltages between the pack and between the two points you called out.

Across pack with pedal all the way down, 37 Volts.

Across M- and B+, looking at a wide range of voltages, from 9V to 23V, but no where near pack voltage.

.

[JohnnieB]

Do you have a stock or aftermarket controller?

If it is a stock controller, the voltage on the white wire to the ITS sensor should read 0.45V to 0.53V when the pedal iis pushed just far enough to make solenoid click.
If so, press pedal slowly to floor and voltage should climb to at least 1.5V.

If so, the controller might be bad.
It could also be high current cables or the F/R switch, but those are new, however new doesn't mean good.....

[cptaylor]

I don't know the full history of this cart, but I am going to ASSUME the controller is the original or a stock replacement. Nothing special, nothing programmable.

The voltage at the white wire (in reference to the negative wire on the battery pack) is 0 sitting. When the pedal is pressed slightly, it engages the solenoid immediately and the voltage goes to 1.5V. As I press the pedal, the voltage will climb up to around 2.5V. Might be higher, I do not have cart jacked up, just trying it on an incline.

It seems to be working correctly right now so I'm gonna have to wait for it to SIA and recheck the white wire again.

just for curiosities sake, how hard is it to take apart the controller and expose the electronics? That's what I do is electronics so if it turns out to be bad, I would rather take a peek before buying another one.

[JohnnieB]

Great, you know electronics.

The throttle input to the controller is the output of a coil whose inductance is varied by how far an iron slug is inserted into it.
Troubleshooting it gets a little tricky because the controller supplies the voltage to the coil and what that voltage is, depends on what controller is used.

The stock controller for a 1996 TXT with a series wound motor is a Curtis 1206 and the source voltage (Pin-2 Black wire) should be slightly above 14V, but less than 15V.

The attached service manual is for a later model year, but EZGO didn't re-invent the wheel every year like the auto industry did, and the only difference is that your cart may not have a reed switch in the charge receptacle.

Section E covers a series controller

------------
As for repairing a controller: There isn't any schematics floating around that I know of, but some people try.

[cptaylor]

Thanks.
Tangent.....
Is there a way to bench test controllers. Will another type of controller work. I have another controller, don't know if it works. Different type. Would like to check it out in my lab, just to see. Was thinking of my power supply, variable, can put out 50 amps, possibly two, and an electronic load or another DC motor. Actually, I've got another golf cart motor also.
Just curious....

[JohnnieB]

The controller type has to match the motor type (Series, Sepex or AC), but other than that, it is just a matter of wiring, voltage and ampacity.

The simplest drive system is the series type, and the controller is basically a DC light dimmer on steroids.
The output of the controller is PWM (Pulse Width Modulated) DC and the throttle input varies the duty cycle of the pulsed DC output, which is pulsing at about 18kHz.

Attached is a manual on a Curtis 1204/5 controller, which is different that the 1206 in suspect is in your cart, but the way it works is the same and the block diagram is close.

The bench test setup would be roughly the same, but the 1206 uses an ITS (Inductive Throttle System), which might be difficult to emulate on the bench without an ITS sensor and the ITS sensor (coil) needed is expensive.

Also, be very careful with cart motors. Check them for wanted and unwanted continuity first.
A bad motor can take out a good controller.

[cptaylor]

So the ITS is nothing more than, say, a toroid wound coil? Possibly a hall effect in there since what I've read about measuring suggest a semiconductor (resistance of a coil in ohms/kiloohms verses megaohms for junction based electronics).

[JohnnieB]

I'm not 100% sure of what it is exactly, but I'm inclined to believe it is just a very long piece of very fine wire wrapped around a hollow tube.
It only has two lead-wires, while Hall Effect devices and similar gadgets have three or more.

Also, the same ITS sensor is used by both stock and aftermarket controllers that have entirely different specs for the throttle input.
IE: The throttle input on my stock Sepex controller (Curtis 1206MX) went from 1.0V when the solenoid first clicked up to 2.9V at WOT while my Alltrax DCX500 goes from 10V down to 6V WOT, both using the original ITS sensor that is installed in my cart.

I'm not sure how one manufacturer managed to get the voltage to increase with the increase in inductance and the other manufacturer got the voltage to decrease with an increase in inductance, but they did. Perhaps one used a constant current source and the other used a constant voltage source.

I suspect it is fed a pulsed DC voltage (probably the same freq as the PWM), which would act like an AC signal and be attenuated by the inductive reactance of the coil.

I've looked at the PWM on a scope, but never thought to look at the ITS when I had access to a scope.