60v to 72v upgrade on a VRLA EVD
I have a 2008 EVD with Lead-Acid batteries. I asked the support guys at R-Martin if the scooter could be upgraded to 72v by just adding another battery, and the answer I got was "Sure - the controller is rated for 80v", though they did say they had never tried this themselves, they had heard of at least one customer who had. So, I added another battery of the same size and type (though it is rated for 40Ah instead of 50Ah), bought a 72v Soneil charger, and charged it up. Now the motor does not come on at all. I bypassed the new battery (downgrading it back to 60v), and the motor works the same as before. The support guys at R-Martin are guessing there may be some overvoltage protection circuit in the controller of which they were unaware. Anyone have any experience with this controller? Anyone know what controller it is? I would love to see the specs on it.
Could you discharge the stock batteries, then add the discharged additional battery? That would rule out any assembly error being the cause of the problem.
If a high voltage cutoff device is causing the problem, then empty batteries might be within the allowed range. You could then charge the batteries up in small steps to find out at what voltage the cutoff happens.
Does your controller look like this one: http://wookware.org/pics/moped/controller/ (which is marked 'BLC3000 Rev 1'). This appears to be OK up to 85V volts or so, but I haven't yet actually tested it out above 54V
I bet it's an over-voltage issue.
What's your pack voltage when fully charged and straight off the charger? If it's 80.x then try just leaving the lights on full beam until it dips below 80V and then give it a try.
Is your new charger tunable at all? It's going to be a PITA to have to leave the lights on... ;-)
I ran the test using headlights to discharge the batteries. (It took uncounted hours to discharge the batteries just a few volts.) The controller would not engage the motor when the battery voltage was 74.7V, but did engage the motor when the battery voltage was 73.6V. I found the circuit I believe is used to monitor battery voltage - Vin goes to R29, which is 140K Ohms, connected to R3 which is 10K Ohms, which goes to ground. The mid-connection is connected to pin 5 of the PIC, which is one of the analog inputs. At 73.6V battery, this point measures 4.9V - The voltage divider circuit divides the input voltage by 15. I plan to change out R29 to something close to 170K (It looks like 165K is the closest standard value) to change the divider to divide by 18 to accommodate the extra battery. The question remains if the rest of the controller as well as the motor can handle the increased voltage. I have asked R-Martin support for the manufacturer and part numbers of the controller and motor are so i can research the specs, but I have not heard back. Not sure if I will hear back.
I doubt R-Martin will be able to get the information - but they may surprise us.
Here's what I'd suggest:
1) You have the controller open so check out the part numbers on the power electronics in the controller (the big capacitors?) and see if you can find their rating. Then you should simply be able to do the math and determine if it's got a chance of handling the increased power. Of course there's other failure points possible but starting there would give you a quick read of yes or no.
2) Even if you do fry the controller you should be able to replace it with a reasonably priced and much more capable Kelly controller. My only pause is not knowing if this is three phase brushless motor you have - I'm pretty sure it is. The other fallback you have is that you can replace the motor and the controller.
3) With increased power comes increased heat. You'll want to consider upgrading the power lines from the controller to the motor and you may also end up needing to replace the hall sensors in the motor (they can be finicky little buggers!)
Of course at some point you might just decide that 5 batteries rather than 6 is "good enough" :-)
Good luck and keep us posted.
The controller uses two identical boards, with a total of 6 different colored heavy gauge wires.
How many heavy gauge wires do you have going from the controller to the rear wheel?
How many fine gauge wires (hall sensor inputs) do you have going from the wheel to the controller?
I'm wondering if your two identical boards / six wires are just doubled up in parallel to provide a 3 phase output. Or if it's something different? (In case you need to get a replacement controller...)
Now, are you sure you wouldn't just rather buy one of mine with 96V? (<-- Just kidding! I'm glad just to offer mildly useful suggestions and encouragement from the sidelines!)
So that's good news - you'll always have a fallback which will be to purchase a kelly KBL controller.
You might actually consider unwrapping that bundle of power wires. We sheath ours with a nylon mesh - anything to get a bit of extra airflow and cooling. The power wires are one of the weakest links in the chain as far as heat durability goes.
Why is it there are always extra pieces left over when you put something back together?
Ahh, yes - a phenomenon that I have experienced myself! Just make sure you don't let the magic smoke out and all should be fine.
Well done on tracking down that other voltage divider. Good luck with your test rides. I think you'll be fine.
I just bought an EVD and was wondering about putting in another cell to jack up the voltage to increase acceleration. Sorry to hear it didn't work.
I was wondering if anyone knew of a desulphator product that works with the EVD battery system?
More power to you...heh.
I have considered all kinds of mods to my stock 3kw EVD also. (1 more batt, aero mods, lighting, etc) However, if I value my time at a reasonable rate, none of them actually pencil out. The stock unit does the job I need done: it carries me from home to work in 28 minutes (24 minutes by truck.)
Particularly given the impossibility of getting technical info from the builder, dinking around has no payback for me. As I have said elsewhere, I am running the charger 15 hours on a day I ride 70 miles, so I will get a 12a charger and plug adapter for the day the 5a charger goes up in smoke. I expect I will have to replace the batts this winter in order to get a full season next year.
The use of my time that does pencil out is researching which fully enclosed 3 or 4 wheeler to buy for commuting in all weather except deep snow. When snow reigns, my 4WD Dodge is hard to beat. (Blasphemous as that is...)
I checked with BB the battery manufacturer. These can take ".3C" charge rate - .3 x 50 - or 15 amps. So a 12 is safe.
Don't know about the search and where I'll post. I like this community so probably here.
I may be the other customer who coverted their EVD VRLA to 72 volts. Perhaps my experience was just dumb luck but I'll share with you what I did to make it happen. First, I ordered an identical battery from RMartin. Second, and upon the battery's arrival, I modified the obvious area for that added battery - level two, under the seat, added to the one battery on the second level. I then added some jumper wires to put the added battery in series with the others and wired the controller in to the series.
I was careful not to wire in the DC to DC controller off the sixth battery but kept it at the fifth one (in series) to avoid frying the converter with voltage beyond 60. For charging, I created my own bank system with six separate 1.5 amp battery maintainers. I had to splice in some added wire to their pig tails but wound up with six separate connectors that are inside the little compartment in front of the seat (locking). When I charge, I just connect each of the chargers to individual pig tails, making certain that each is protected sufficiently. The pig tails are protected with caps when not charging due to exposed ground wires for individual batteries that could connect with others and spark up a storm. This system essentially gives me identical votages for each battery and a mechanism to tell me if one of the batteries is failing. BTW, there is about a five minute difference between when the first battery reaches optimal voltage and the last one does. That merely tells me that the system is working. I rarely let the batteries get below half.
I weigh about 225 (shouldn't weigh that much but that isn't the point of this forum). I made 50 mph on the flat when the EVD ran on 60 volts. It now gets up to 55, taking a long time, as expected. The thrust off the line is scary actually and I have to tell would-be test drivers to take it easy on the grip to start with. The standard controller is pretty meager, in my estimation, and evidently uses lightswitch type values for inaugural pulse width modulation. I suspect that other controllers are much better (see below).
Now, I have the same problem with the thing overheating and just about dying. If it is a oool day, I am fine. Anything in the seventies or more and the scooter will slow down to a 5 mph crawl for about 100 yards, then start chugging on and off until the controller appears to reset and I take off as it nothing ever happened. When I get back to home after this occurs, the hub motor is very hot but hte controller seems to be about normal. My first controller lasted less than a mile before it ws shooting flame-style aparks out of it. RMartin sent me a new controller and that has worked fine.
Hard to know what is causing the slow-down to a crawl. With the lights on, the bike tends to surge a bit. I'll be riding at 35 mph and the thing then starts to slow down to 32 or so and it takes off again. I tend to add juice when this happens and suspect that it may be related to the overheating.
I'm now thinking about changing to a Sevcon Sepex controller. They have an 80 volt unit that is available through Electricmotorsports. One of my concerns is that the EVD wiring diagram isn't as descriptive as I'd like (and I pointed out a mistake in it to the factory and they concurred.) I'd like a breakdowm of the actual deployment of each color wire so that I don;t have to take the controller apart and figure out what goes where. Perhaps you have that down since you've been into the innards a bit more.
Anyway, just wanted to weigh in on this. Not sure how much help I've been. If you have some ideas about the overheating (that occurred equally under 60 volts and 72) I'd enjoy knowing about it.
Thanks for the detail. I, too, found the nylon sheathing around the bolted connectors and also felt that the route that the wires had to follow likely places a material strain on the loom. They were right to use different lengths of wire before the bolts, making it less likely that a short could occur. But, with all the twisting etc., such is bound to happen.
As far as Kelly Controllers, consider the following string, amongst many:
I do want to upgrade my controller but am leaning toward the Sevcon Powerpak. There's stateside backing from a couple of distributors of these UK-built controllers.
Anway, I am grateful for your helpful comments and wish you well in this project. Please keep us updated on your progress and I will post more as my project moves along as well.
Thanks for the info on the motor type. I had actually been considering changing out the motor and controller with a matched Kelly pair but the postings on customer service have made me think again. If you find a controller that works, please let me know. I am a bit frustrated with the slow-down to 5 mph routine - particularly if I'm in the left lane of a city four lane.
I get the two, followed by three blinks on the LED and that makes sense since the unit slows down to 5 mph. I presume that it would stop altogether if the brake circuit was activated by the sensor from an overheated motor. What is interesting is that it eventually chugs a good bit before resuming full speed operation. Perhaps the temperature sensor in the controller is approaching threshold values as it cools off.
When I get home, the controller seems warm to the touch and the hub area is hot. I've never checked the brake lights but will. The issue for me is that it has rarely stopped completely so either the motor temperature situation is within tolerances or the sensor is not working.
I'll ponder the motor and controller cooling idea. Somehow an airflow has to be established that doesn't expose the innards to the elements.