2021 icon i40 with Roy pow lithium battery locks up

[fixed4life]

Originally Posted by ATrain

Your #2 sounds like what I was thinking from the beginning, except your suggestion avoids an auxiliary power supply all together. I haven't worked with lithium enough so I was unaware you could bypass the BMS.

I'd actually do 2 relays for redundancy and have them with basic logic where the keyswitch on + BMS off = EMB off. Otherwise, EMB = on, except the BMS/controller would work as normal turning the EMB on/off as needed.

That way, in the event of a BMS event, the loss of power keeps the EMB on. Essentially turning the keyswitch into an auxiliary tow switch.

I was also thinking about the open circuit triggers bypass relay arrangement, but my concern was that I do not know what the effect of "X" ms of a lapse of current could be.

[ATrain]

Originally Posted by fixed4life

I was also thinking about the open circuit triggers bypass relay arrangement, but my concern was that I do not know what the effect of "X" ms of a lapse of current could be.

I had that same concern! The easy solution is just configure it in a NC (normally closed) configuration so there (1) is no delay and (2) relay failure just means your EMB never closes...easy to detect and resolve.

Even if you did a NO (normally open) configuration, I think the main concern is not how many ms to close the relay, but how many ms for the EMB to lose enough power for the electromagnet to stop holding the brake open AND for the springs to push it closed.

It's essentially small spring + small gap vs large spring + larger gap + time for electromagnet to weaken...I'd think anyway.

[dundeebarnbuggy]

Thinking outside the box here. I don't know who makes the Icon, but wouldn't it be easier [and safer] to get the parts from an older model to put back the conventional parking brake on the pedal? Then disable the electric one.

[fixed4life]

Originally Posted by ATrain

I had that same concern! The easy solution is just configure it in a NC (normally closed) configuration so there (1) is no delay and (2) relay failure just means your EMB never closes...easy to detect and resolve.

Even if you did a NO (normally open) configuration, I think the main concern is not how many ms to close the relay, but how many ms for the EMB to lose enough power for the electromagnet to stop holding the brake open AND for the springs to push it closed.

It's essentially small spring + small gap vs large spring + larger gap + time for electromagnet to weaken...I'd think anyway.

I am probably missing something, but it seems to me that even if you run NC and it is in the open position, the small spring + small gap (potential) problem would still exist, no?

Whichever way the OP goes, the PSA here is that people need to get the EMB pre-BMS if you are going to run one of these batteries of unknown internal specifications. Those batteries are designed to protect themselves, not you and tend to use the minimum specification components that will allow the ratings listed. The damage that can occur when something goes wrong is far greater than potentially voiding the warranty of the battery.

After this exercise (thank you for the brainstorming ATrain) I believe my plan when I move my chassis wiring to post-BMS ground on my battery (although my setup is different, the issue would then become the same) is to isolate the EMB and use a dash switch, in conjunction with an independent speed sensor / relay so that the dash switch can not be engaged above 0 MPH. The more I think about it, the more I like the idea of a button for the parking brake actually (I use my cart around town, not for golf generally). Maybe I will finally start documenting these projects...

[ATrain]

Originally Posted by fixed4life

I am probably missing something, but it seems to me that even if you run NC and it is in the open position, the small spring + small gap (potential) problem would still exist, no?

Nope (AFAIK). And for my idea, I'd have to sit down and draw the circuit out with multiple relays and think it out, but I think it can be done. I also could be completely wrong, but I think I have it. Those relays can be wired 2 ways (NO/NC), and I'm suggesting the less common way (NC).

What you're saying is when the BMS cuts, the tiny electromagnet in the relay should activate/close and then provide power to the EMB. That tiny delay might be a problem. That's a normally open relay. (side note, I think it might still be faster than the EMB can close even wired this way).

What I'm saying is normally closed. Where constant power is always provided to the EMB, so there is zero delay. To activate the EMB, voltage must be provided to the relay, which activates the little spring/magnet, and then cuts power to the EMB. Somewhat counterintuitive to provide power to cut power.

The KEY here how this could work is literally the keyswitch. This eliminates your need for a speed sensor. Without the keyswitch, when you turn your cart off, a small amount of constant power would be needed to keep the relay open and your EMB closed. That would very slowly drain your battery, which likely would be negligible, and even still a dead battery ultimately closed the EMB. The actual risk would be if the relay (spring) failed, then your EMB would open and cart could roll away.

We can avoid that though with the keyswitch! By putting the keyswitch before the relay circuit that always provides power...when you turn off the key, it cuts power to the relay so both channels of power to the relay are cut, thus normal EMB behavior. And the beauty of this solution is if your relay failed, your EMB would just remain open at all times the key is on, which is far safer than it randomly closing.

Since you're actually going to do this, I want to find some time to draw out the circuit and see if we can come up with a permanent solution, because an e-brake button sucks. It doesn't sound that bad until you actually experience the cart without it and realize how often the EMB actually engages.

I only know this because I was stupid enough to tighten those 2 screws on the EMB when I saw they were loose and I drove my cart around for a couple weeks. I suggest you try that yourself first and see how weird it is driving the cart.

Waiting at a stop light on a slight incline meant I had my foot mashed on the brake and when I went to drive forward, the cart would roll slightly back towards the vehicle behind me during the delay for the motor to kick in. Parallel parked on a hill, the cart would roll all over the place for tiny fractions of a second while I'm going forward/backwards. Parallel parking was harder. I had to use both feet for both pedals sometimes. One perpendicular parking spot was virtually flat and my cart was stationary, but when I got out of the cart and walked away, I looked back and it was slowly rolling out...so if you ever forget to push the button lol. Granted I live in hilly CA, not FL.

[fixed4life]

Originally Posted by ATrain

Nope (AFAIK). And for my idea, I'd have to sit down and draw the circuit out with multiple relays and think it out, but I think it can be done. I also could be completely wrong, but I think I have it. Those relays can be wired 2 ways (NO/NC), and I'm suggesting the less common way (NC).

What you're saying is when the BMS cuts, the tiny electromagnet in the relay should activate/close and then provide power to the EMB. That tiny delay might be a problem. That's a normally open relay. (side note, I think it might still be faster than the EMB can close even wired this way).

What I'm saying is normally closed. Where constant power is always provided to the EMB, so there is zero delay. To activate the EMB, voltage must be provided to the relay, which activates the little spring/magnet, and then cuts power to the EMB. Somewhat counterintuitive to provide power to cut power.

The KEY here how this could work is literally the keyswitch. This eliminates your need for a speed sensor. Without the keyswitch, when you turn your cart off, a small amount of constant power would be needed to keep the relay open and your EMB closed. That would very slowly drain your battery, which likely would be negligible, and even still a dead battery ultimately closed the EMB. The actual risk would be if the relay (spring) failed, then your EMB would open and cart could roll away.

We can avoid that though with the keyswitch! By putting the keyswitch before the relay circuit that always provides power...when you turn off the key, it cuts power to the relay so both channels of power to the relay are cut, thus normal EMB behavior. And the beauty of this solution is if your relay failed, your EMB would just remain open at all times the key is on, which is far safer than it randomly closing.

Since you're actually going to do this, I want to find some time to draw out the circuit and see if we can come up with a permanent solution, because an e-brake button sucks. It doesn't sound that bad until you actually experience the cart without it and realize how often the EMB actually engages.

I only know this because I was stupid enough to tighten those 2 screws on the EMB when I saw they were loose and I drove my cart around for a couple weeks. I suggest you try that yourself first and see how weird it is driving the cart.

Waiting at a stop light on a slight incline meant I had my foot mashed on the brake and when I went to drive forward, the cart would roll slightly back towards the vehicle behind me during the delay for the motor to kick in. Parallel parked on a hill, the cart would roll all over the place for tiny fractions of a second while I'm going forward/backwards. Parallel parking was harder. I had to use both feet for both pedals sometimes. One perpendicular parking spot was virtually flat and my cart was stationary, but when I got out of the cart and walked away, I looked back and it was slowly rolling out...so if you ever forget to push the button lol. Granted I live in hilly CA, not FL.

I am familiar with running relays that way, perhaps I will diagram what I think you are suggesting so we can compare notes. I must be missing something.

I guess my concern with such a short blip of current loss really stems from my lack of understanding about how our EMB works. If nothing physically engages (i.e it is like a door magnet) I imagine it would be no problem, but if something does, I imagine the force at play would prevent any hope of disengagement. Do you know what the engagement mechanism looks like?

You make very good points about the manual parking brake, and paint quite a picture... I may need to rethink my position on that one.

[Pat911]

You need to consider battery safety as well. Connecting the EMB prior to the BMS will mean that the EMB magnet cannot be disconnected to protect the battery from being over discharged. This needs to be considered when designing any solution to this problem as even a single incident of a cell getting below it’s threshold voltage will render it unsafe to use.

An acceptable solution would be for the BMS to cut power to the controller, allowing the cart to come to a gradual stop and then removing power to the EMB 30 seconds later. 30 seconds will give plenty of time for the cart to stop.

[ATrain]

Originally Posted by fixed4life

I am familiar with running relays that way, perhaps I will diagram what I think you are suggesting so we can compare notes. I must be missing something.

I guess my concern with such a short blip of current loss really stems from my lack of understanding about how our EMB works. If nothing physically engages (i.e it is like a door magnet) I imagine it would be no problem, but if something does, I imagine the force at play would prevent any hope of disengagement. Do you know what the engagement mechanism looks like?

You make very good points about the manual parking brake, and paint quite a picture... I may need to rethink my position on that one.

This video shows pretty well how they work - https://www.youtube.com/watch?v=Sn5iIya2qNI

I'm over explaining some of this stuff for other readers as I know you understand relays, but other people may not, so don't think I'm patronizing you.

It's basically a clutch plate. I think essentially on the end of the motor drive shaft is a disk with teeth that the EMB presses down on. It has springs that push it down at all times. It is magnetic too and when you provide power to the EMB, the electromagnet turns on and pulls the brake plate/clutch off the disk at the end of the motor drive shaft and allows it to spin freely.

(1) I would guess the relay delay is less than the time it takes for the electromagnet to release and the springs to close that gap, because it's a further distance and a heavier object for the sprints to push. And (2), when the electromagnet loses power from the BMS, it doesn't immediately become demagnetized as well, so there are probably a few milliseconds gained there as well.

This makes me think even a normally open (NO) relay would work to prevent the EMB from engaging, but it would still need a keyswitch or speed-sensor controlling it. I had originally thought of the NO relay and had the latency concern and relay failure concern. If the relay fails and closes, you're getting launched. My plan was to have 2 relays for redundancy and hook them to dash status lights so you could always verify your relays are working and safe.

After thinking harder on it, that's when I realized everything could be wired to 1 NC relay and there would be virtually no safety concerns. A relay failure would merely mean your EMB does not engage unless you turn the key off. A BMS failure and a relay failure would still keep EMP open.

A total loss of all power (cutting battery cable?) or turning the key off mid-drive would be the exact same as it is currently.

Turning the key off while driving is still an unanswered question for me...but this idea has no impact either way on it.

Next week I plan to try and draw it out sometime. I could end up getting to the end of my drawing and realizing I'm an idiot, so hopefully I won't look too foolish if that turns out to be the case.

If my line of thinking is correct, I was going to try and produce a "kit" device. I think the ICON EMB's have some sort of automotive plug on them. I'd try and find that same plug and get a coupler of some sort and that type of thing so people could DIY it or dealers could easily install it.

I really want lithium at some point but I even more want it to be safe. I like to cruise 35mph+ and sometimes I have a few drinks on my cart. I can't risk having it randomly do a barrel roll lol.

[ATrain]

Originally Posted by Pat911

You need to consider battery safety as well. Connecting the EMB prior to the BMS will mean that the EMB magnet cannot be disconnected to protect the battery from being over discharged. This needs to be considered when designing any solution to this problem as even a single incident of a cell getting below it’s threshold voltage will render it unsafe to use.

An acceptable solution would be for the BMS to cut power to the controller, allowing the cart to come to a gradual stop and then removing power to the EMB 30 seconds later. 30 seconds will give plenty of time for the cart to stop.

Absolutely this. In my original comments I had mentioned using one of those USB battery banks with pass-through-charging as an auxiliary power source instead of the pre-BMS connection.

The ideal solution is proper BMS to controller communication, but the issue is RoyPow or whomever makes the lithium packs to be used as a drop-in for various carts...so I doubt they'd handle that. And they probably have a variety of BMS devices they use depending on what is in stock in China and cost. Plus the various controllers across the different carts.

I spoke with people at ICON HQ and they said lithium is on their agenda, but it would be a factory offering.

So the only solution I see is the pre-BMS power and/or auxiliary power source. With the normally closed relay and the aux power source, I think that could be a perfect solution because then you have no risk to the lithium pack. There would need to be some aux battery status light somewhere though. If that failed AND your BMS cut out, then you'd be in trouble.

[skrause]

Had a very similar issue today with our i40FL. Our dealer installed evolution 110ah lithium battery when we initially bought the cart in September 2021. We always have it charging when not in use. Today, it was at 100% charge when we left the house. We drove about 1 mile and stopped for about 45 minutes and parked. Came back and drove for about another 1 mile. Turned onto a busy road and doing about 20+ mph the cart locked and came to a skidding halt. Almost pushed us into oncoming traffic and we almost flipped. It pulled to the left. Once it stopped we were fortunate to avoid being rear ended but we were stuck in the middle of the road for at least 10 minutes because the brakes locked and we are were unable to push or get to neutral. We were told by our dealer to hold the reset button for 15 seconds. At that point the cart started beeping and we were able to actually drive it to the side of the road but the battery meter said 0%. It was over 90% just a few seconds before the incident. Similar conditions. 65-70 degrees outside. Our dealer came and towed the cart and we are awaiting next steps. We had no idea this was possible and consider ourselves lucky to have avoided what could have been catastrophic. Now we are obviously terrified of driving it again. Especially in any sort of busy roads. We are in CA and ours is street legal. These carts have no business being on any road with the brake locking being a possibility.