XM-3150 Throttle Response

23 posts / 0 new
Last post
Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
XM-3150 Throttle Response

XM-3150 Throttle Response

I thought I’d start a new topic for this issue. In my previous review of this scooter I noted an issue with the throttle response - I thought it was abrupt especially at mid throttle cruising speeds. I did some investigating and found that when you plot signal voltage over twist grip rotation, the curve has an “S” shape and is indeed steep in the middle.

To make my measurements I fastened a protractor on the end of the twist grip and made a pointer. Then I removed the left side cover to gain access to the ‘economy’ switch, where I was pleased to find that the wiring was very easy to manipulate thanks to a jumper harness that the factory added just to accommodate the economy switch. The circuit only involves 3 wires - 5 volt power supply, on one end (usually red), a ground on the other end (usually black), and the signal return in the middle (white or gray). You can simply unplug the harness from each end and plug the remaining connectors together to bypass the switch entirely. All the jumper harness does is allow the economy switch to add a 330 ohm resistor in series to the 5 volt power supply, dropping it to 4 volts. But then you can take that unused jumper harness, add a tap to each of the three wires, and make your very own ‘break out box’. Just get a voltmeter and a bundle of those multi-colored jumper test leads and measuring is fun and easy.

First I measured the stock twist grip in the ‘power’ mode, taking voltage readings every 10 degrees. Then I entered the data into a “works” spreadsheet and let it make a graph. (By the way, make sure the bike is on the center stand with the rear wheel off the ground and that the side stand is down - the motor will run if the side stand is up.) First observation - the grip has 70 degrees available, but the first 25 degrees are not being utilized. The sensor returns .8 volts at zero drgrees and does not reach the motor start thresh hold until right around 1.4 volts at just over 25 degrees,. At the other end, the voltage maxes out at 4.2 volts at 60 degrees - leaving another 10 degrees unused. That means that you have only 25 degrees to work with which is about 1 / 16 of a turn - which is not nearly enough if you ask me. In the middle of the range, the slope is steep, gaining a volt for every ten degrees., twice as much as a straight line would be from beginning to end.

I hooked up the econo switch and plotted it - it looks much the same as the ‘power’ graph although not quite as steep and signing off at 60 degrees with 3.2 volts which equals 30 mph on the road.

Next I rigged the protractor to a 5000 ohm potentiometer. These are cheap and easy to come by at an electronics store. As expected, the graph was a nice straight line but the only problem is the pot uses 270 degrees of rotation. To hook one up one would probably need an old fashioned throttle cable leading to a remote Rube Goldberg device. I also tried a 1000 ohm pot thinking it might have a different slope but it plotted almost exactly the same as the 5000 ohm one did. I guess they return a percentage and it doesn’t matter if its 1000 or 5000.

That’s where it stands at the moment. I need to plot some other hall effect sensors and see if they’re all the same. On the potentiometer side I’d like to try adding a resistor to the ground side and see if that affects the slope. If I can adjust the slope with resistors then I could tune the start and end points mechanically with my yet to be built “device”. If anyone has any questions, suggestions, or advice - please don’t hesitate to chime in.

LeftieBiker
Offline
Last seen: 2 years 9 months ago
Joined: Saturday, July 9, 2011 - 04:36
Points: 886
Re: XM-3150 Throttle Response

I'd look at different throttles from different, but similar, bikes. My ZEV has a similar problem - too abrupt a transition from not enough power to just a bit too much at low throttle. My XM-3000 doesn't seem bad to me - maybe you should try one of those throttles. While you're at it, get the switch set with the Economy/Power switch on it, since you seem to be able to wire it.

mf70
Offline
Last seen: 6 years 8 months ago
Joined: Friday, December 1, 2006 - 09:01
Points: 712
Re: XM-3150 Throttle Response

Thanks for the complete write up. It will be interesting to see how much variation there is between various throttles. Your description will make it easy for anyone to record their throttle response. It'll be a few weeks before I can do it, but I'd be happy to contribute when I can.

Mark

LeftieBiker
Offline
Last seen: 2 years 9 months ago
Joined: Saturday, July 9, 2011 - 04:36
Points: 886
Re: XM-3150 Throttle Response

When I took my XM out for a ride yesterday evening, I paid attention to the throttle response, and actually it seems to have the same profile as yours. Nothing for the first 25%, then a slightly jerky (in High, anyway) surge, then a rapid increase in power to the end of the range, which isn't very far. I'm just used to it, I guess. Heck, I was already getting used to the ZEV's much worse off-stop response on my second ride...

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

Good news on the throttle response front. All hall throttle sensors are not created equal. I received a hall effect throttle from TNC Scooters today (part #101100) and quickly ran a plot. It was a lot different than the stock one. The TNC has only 65 degrees of total rotation (5 less than stock) but it doesn’t waste any in initial opening, rising from .83 at idle to the motor thresh hold of 1.4 volts in just the first five degrees. It only takes 5 more degrees to reach almost 2 volts. Then it begins to level out, taking the next 30 degrees to reach 2.7 volts, before climbing steeply again to 4.2 volts at 60.

The graph suggested the bike would come on suddenly, have a big soft spot in the middle, and have a kick at the end. I installed it without too much trouble and took it for a ride. Initial observation was a huge relief - my scooter doesn’t jerk anymore! I did some more riding and found that the plot is a very good prediction of how the throttle will operate. This TNC throttle is a big improvement over stock, however it is not perfect. The soft spot in the middle is great for putting around the neighborhood at 15 - 25 mph but the last third is still too steep and some jerkiness returns around 30 to 40. On the plus side, it has some not very accurate but useful graduated scale markings to indicate throttle opening and the thing only cost 9 bucks. You can even use your original end cap.

The quest goes on for the perfect throttle. I saw one on the Kelly Controller web site for $15. I think I’ll try that one next. I’m not sure how to imbed charts. Let me see if I can tabulate my data so far:

Rotation Stock Econo POT TNC
0 0.82 0.83 0.69 0.84
10 0.86 0.86 0.87 1.97
20 0.94 0.92 1.05 2.36
30 1.70 1.40 1.22 2.52
40 2.76 2.30 1.40 2.67
50 3.90 3.10 1.60 3.10
60 4.2 3.2 1.95 4.28
70 4.2 3.2 1.95 4.28

Charles

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

Throttle update - The TNC throttle has been doing good - I can live with this one. I was going to order a Wuxing brand throttle from Kelly but they wanted $23 to ship the $15 unit ($38 total) so I balked. TNC only charged $6.05 to ship a $9 throttle by USPS priority mail. Also, when I took the stock throttle off I noticed it was a Wuxing, so I’m a little wary of that brand. I might order another TNC and see if it plots the same. It’s still not clear whether they vary individually or by brand.

Charles

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

Throttle Response Update #2
Only because they’re so cheap ($9), I ordered two more throttles from TNC - one just like I had (we’ll call it TNC2), and another ½ twist one (TNC½). To speed things up, I installed common (but insulated) female blade connectors near the handlebar and put the males on the throttle wires. I hooked up the TNC2 and ran a plot. I was a little disappointed to find that it was almost identical to TNC1. While TNC1 was a lot better than stock I was hoping for a more linear response. Oh well, at least that suggest that (based on only two samples) that identical throttles sometimes behave the same.

I took the ½ twist one out of the bag and was not impressed. It was similar but not identical to TNC1. It looked, felt, and operated cheap. It had a lot of ‘stiction’, so much so that it wouldn‘t always return to idle. I tried adjusting it (prying the trim rings away from the grip, etc.) but there was no difference. Some stiction is not necessarily bad. The stock Wuxing throttle had no stiction whatsoever and that‘s one reason it was so hard to control over bumps. TNC1 had what I considered to be the right amount. Anyway, I didn’t have much hope for it (TNC½) but I plotted it anyway.

It seems they all start out at 0.83 volts at idle and this was no different. At 10 degrees it was still under the motor start thresh hold, but to my surprise it tracked a straight line the rest of the way up to 60 degrees. Hmmm, this thing might be worth a try. I took it for a ride and sure enough, I think it’s an improvement over TNC1. I really wasn’t fond of ½ twist throttles but I really like this ones linear response. The stiction is still there, so I don’t know that I can really recommend a throttle that doesn’t return by itself. But for me, it was no problem at all - almost a cruise control feature even.

My conclusion is this - if you’re having a hard time regulating your speed, so much so that it constantly irritates you, you don’t have to live with it. Try some different throttles. And let us know if you find some good ones.

One other question. It’s my understanding these hall effect sensors are transistor based and thus will never be able to return the full 5 volts supply to the signal return because it always takes 0.7 volts to turn it on. Are we leaving anything on the table? Will it go faster if it sees 5 volts instead of 4.2? I guess I could wire up the POT and see for myself.

One last tip. Remember in my review I remarked that the rear brake was more powerful than the front? I found out why. The front brake lever was in need of lubrication, both at the pivot and especially where it rubs the master cylinder valve. A little bit of grease at this critical point increased my front brake power (or reduced the effort anyway) by at least two times.

Charles

mf70
Offline
Last seen: 6 years 8 months ago
Joined: Friday, December 1, 2006 - 09:01
Points: 712
Re: XM-3150 Throttle Response

One reason a sticking throttle might not be as horrible as a car's is that the brakes are interlocked to the controller.

MF

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

Very interesting discussion to me. I just bought a ZEV 7100 Alpine and the throttle response is very jerky, with a huge power difference over a very small change in throttle position. I hope to be contributing more information later, and may be trying the TNC throttle. Thanks for your research!
- Alan

ZEV 7100 Alpine
Fort Collins, CO

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

I've added the ZEV throttle response to your data, and put it into .csv format so it's easy to import into excel - I rearranged the columns because I was not very interested in the Econo or POT columns. The ZEV throttle is actually pretty linear, despite my subjective experience of it's behavior. I've ordered a TNC throttle to see it it improves things - it seems to be pretty flat in the throttle range where I usually cruise. It's getting it to cruise smoothly that I have the most trouble with - usually around 40 - 45 mph for my travels.

Rotation, Stock,ZEV, TNC,, POT, Econo
0,0.82,0.847,0.84,,0.69,0.83
10,0.86,0.95,1.97,,0.87,0.86
20,0.94,1.42,2.36,,1.05,0.92
30,1.7,2.05,2.52,,1.22,1.4
40,2.76,2.57,2.67,,1.4,2.3
50,3.9,2.95,3.1,,1.6,3.1
60,4.2,3.5,4.28,,1.95,3.2
70,4.2,4.25,4.28,,1.95,3.2
80,4.2,4.58,4.28,,,3.2

Other notes:
I get about 80 degrees of rotation on the ZEV - I just copied the 70 degree value to extend the range for non-ZEV data.
The throttle input voltage on the ZEV is 5.3v.

- Alan

ZEV 7100 Alpine
Fort Collins, CO

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

Thanks for the data astar.
I ordered a Magura 5K POT throttle ($50 plus shipping). It should arrive this weekend. I'll let you know how that goes.
In the meantime I've gone back to the TNC1. On a side note I mounted a Doc Wattson (a cheaper version of the Cycle Analyst). I like it, but I need to mount it and wire it better.

Charles

mf70
Offline
Last seen: 6 years 8 months ago
Joined: Friday, December 1, 2006 - 09:01
Points: 712
Re: XM-3150 Throttle Response

Now that Y'all have extra throttles, it's time to start taking them apart to see if they can be modded for more progressive response. The variable part is a curved magnet that is brought alongside a Hall-effect sensor.

Mark

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

The Magura throttle came in today. Instead of plotting it first, I put connectors on it and took it for a ride. It was nice and linear, but it had a ‘lot of slop in the throttle cable’. In other words it didn’t move until you twisted the grip quite a ways. I expected that, being as how this is a 0-5000 ohm potentiometer, not a hall effect. I brought it in and ran a plot. This twist grip has about 73 degrees available and it tracks pretty much a straight line from 0 to 5 volts. That means it wastes 26 degrees getting to the motor start threshold of 1.45 volts and another 10-13 degrees after arriving at 4.2 volts wide open. That’s 78 millivolts per degree, not that much better than the stock throttle which returns 98 millivolts (in the meat of the ‘power band’ - from 20 to 50 degrees).

But with a potentiometer, you can fine tune the start and stop points with resistors. I went to Radio Shack and bought an assortment pack of ¼ watt ‘carbon film’ resistors (about $7). First I added a 1000 ohm resistor in series to the ground wire and sure enough, it bumped the start point up by about .7 volts and ‘shortened the throttle cable’ by about 10 degrees. Then I added another 1000 ohm resistor to the red 5volt supply wire. That bumped the end point down from 5 volts to 4.3 (still wide open but at 70 degrees, not 60).
I changed the ground resistor from 1000 to 2000 ohms and that improved the starting point up to 1.3 volts, still about 8 degrees away from the motor start threshold. At these settings the slope was 46 millivolts per degree - about half as much as stock and straight as an arrow. The TNC throttle had a slope of 24 millivolts in the mid-range, no jerking but that was aggravating in its own way - you twist the throttle and nothing happens. Then at the end it jumped up to about 90 millivolts from 50 to 60 degrees.

I tidied it up and took it for a ride. Throttle response off the line is very good. Mid speed is good. Faster speeds are good, but it tends to be bumpier at faster speeds. Its especially better when resuming throttle after a turn. At long last, Throttle Nirvana. Well, just about. There are still some issues. The Magura doesn’t swing the cable out far enough to clear the switch housing so I still need to do something about that. Next I'm thinking about adding another battery.

Charles

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

I put the TNC throttle on my ZEV today, and the throttle response is much improved. It no longer has that on/off switch feeling when trying to maintain a steady speed. It also starts responding with less throttle opening as your data suggests. I think this is good enough for me - I'm pretty happy with the response. Thanks!

ZEV 7100 Alpine
Fort Collins, CO

LeftieBiker
Offline
Last seen: 2 years 9 months ago
Joined: Saturday, July 9, 2011 - 04:36
Points: 886
Re: XM-3150 Throttle Response

Can you give me a url for the TNC throttle? Maybe next year I'll get one for my own ZEV. I don't find the throttle response dangerous or worrisome but I sure wouldn't call it "smooth," either.

IBScootn
Offline
Last seen: 6 years 9 months ago
Joined: Monday, January 3, 2011 - 14:56
Points: 257
Re: XM-3150 Throttle Response

tncscooters.com part number 101100

Motorcycles: 2011 ZEV Trail 7100, 84V, 60AH, 60+mph, Cycle Analyst, TNC throttle, modified charger. 2013 Kymco GT300i
Bicycles: 2017 Sondors Thin
Cars: 2016 Leaf SV, 30KWH pack. 2007 CR-V
Solar array: 5KW. Cost per lifetime KWH produced $0.073
Bi

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

For the ZEV, here is how the wires should be mated to the TNC throttle:

-----------ZEV Harness---TNC
Ground-----Black---------Black
5 volt-------Pink----------Red
Throttle-----Red----------Green

Good luck and let us know how it turns out.

ZEV 7100 Alpine
Fort Collins, CO

LeftieBiker
Offline
Last seen: 2 years 9 months ago
Joined: Saturday, July 9, 2011 - 04:36
Points: 886
Re: XM-3150 Throttle Response

Thanks for the site name, part # and wiring matchup. Maybe I'll order it now, just in case...

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

Astar, your ZEV throttle looks pretty good on the graph - very linear with a slope of only 56 millivolts per degree. You might think about adding a resistor to the ground wire of that stock control. That should bump up your starting voltage and give a wider shallower graph. I would try something in the range of 100-200 ohms and see how that plots. With my stock hall throttle my econo resistor measured 330 ohms and dropped the high end by about a volt. Does your ZEV have a "transmission"? I see where some of those bikes have speed range buttons. I wonder how those are implemented. Does the throttle plot diffently in different gears? Do they just switch in resistors to limit speed? That's sort of what they did on my "two speed". Anyway I'm glad to hear you like your new throttle. I'm in love with my Magura

Charles

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

Hi Charles,

The ZEV does have a 3 speed transmission, but I don't think it's as simple as the High/Low switch on the XM. My graph above was plotted in 1st gear. The gears on the ZEV are current limits within the controller. I don't really understand the details but it is modifying controller parameters on the fly I think.

Although I really like the TNC throttle, I think your suggestion is good; thanks. If I decide to do it, I will buy two 500 Ohm 10 turn potentiometers. One for the 5v line and one for the ground side. That way I could tune the throttle movements to be in the optimal range of the controller - bring up the bottom to be just before the motor engages, and dropping the top so I get all the speed, but don't waste throttle movement on unusable voltage changes. I'll put these pot's upstream of the throttle so that I could apply them to any throttle I install, and I'll mount the pots so that I can get to them without taking off too many panels. After using the TNC, I'm not convinced that a perfectly linear throttle is what I want.

I'm glad you like your Magura. It's not a hall effect, it's resistive, right? I wonder if there is a down side to resistive; they must use hall effect for a reason. I would think that the hall effect would be slower to wear than a resistive throttle. Let's see how your Magura does over time.
- Alan

ZEV 7100 Alpine
Fort Collins, CO

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

I did the experiment using the stock ZEV throttle to see if I could improve the throttle response. I got only a small subjective improvement, and the data shows why. First the setup: I put a 1K Ohm trim pot (15 turns) on the ground line and the 5v line between the throttle and the controller. The ground resistor allows me to pull up the start voltage, and the 5v resistor allows me to pull down the maximum voltage at full throttle. I refined my data collection to get more accurate throttle openings than last time (I was pretty sloppy up above). I took data with 4 settings:

Stock ZEV:
Ground-- 0 Ohms,
5v line- 0 Ohms,
Throttle Range: 0.855v to 4.54v

ZEV-1
Ground-- 147 Ohms,
5v line- 0 Ohms,
Throttle Range: 1.34v to 4.55v

ZEV-2
Ground-- 157 Ohms,
5v line- 194 Ohms,
Throttle Range: 1.35v to 3.94v

ZEV-3
Ground-- 166 Ohms,
5v line- 289 Ohms,
Throttle Range: 1.355v to 3.73v

The motor seems to kick in around 1.6 volts, but I could not get closer on the bottom end because the controller will not turn on unless it sees a voltage below about 1.36v; this is a safety feature to make sure it doesn't have an engaged throttle when the key is turned on.

ThrottleCurve.jpg

Here's the data:
Rotation,XM-Stk,ZEV-Stk,ZEV-1,ZEV -2,ZEV-3, TNC
0,0.82,0.855,1.35,1.34,1.345,0.84
10,0.86,0.9,1.4,1.39,1.38,1.97
20,0.94,1.65,1.92,1.87,1.71,2.36
30,1.7,2.17,2.49,2.27,2.16,2.52
40,2.76,2.74,2.95,2.68,2.58,2.67
50,3.9,3.37,3.58,3.18,3.07,3.1
60,4.2,4.24,4.29,3.81,3.66,4.28
70,4.2,4.54,4.55,3.94,3.73,4.28

The XM-Stk and TNC data is from Chasbro.

If you plot these, you will see that the slope of each line in the 20 degree to 50 degree throttle range is not that different (except for the TNC). In fact, the absolute value is not that different either because the stock setup jumps up quickly a the beginning. Anything happening below 1.6 volts doesn't make any difference since the motor won't turn. There might be some control advantage in the upper throttle openings for some of the setups, but that would be for cruising at highway speeds I think. I'm not sure if reducing the top voltage is limiting the maximum controller output or not.

I think I'll switch back to the TNC.

ZEV 7100 Alpine
Fort Collins, CO

Chasbro
Offline
Last seen: 11 years 3 months ago
Joined: Thursday, July 14, 2011 - 19:13
Points: 38
Re: XM-3150 Throttle Response

Thanks again for the info Astar,
I read up a little on your Alpine. That thing must be a beast. I can see how you might need a soft spot in the middle to tame that thing. Yeah, I might have a wear issue with the Magura. At work (I’m a bus mechanic) I deal with some throttle sensors (Navistar calls them ‘acceleration position sensors‘). Like the Magura, they have a potentiometer driven by little plastic gears. The problem is they wear and quit agreeing with the Idle Validation Switch (Navistars safeguard against runaway vehicles).

Charles

astar
Offline
Last seen: 10 years 5 months ago
Joined: Friday, January 11, 2008 - 22:11
Points: 158
Re: XM-3150 Throttle Response

And finally, the TNC throttle with no resistors, and tuned to reduce the voltage range; feels pretty good either way:

TNC_Throttle.jpg

ZEV 7100 Alpine
Fort Collins, CO

Log in or register to post comments

Use code"Solar22" and enjoy 12% off for all solar Kits.


Who's online

There are currently 0 users online.

Who's new

  • eric01
  • Norberto
  • sarim
  • Edd
  • OlaOst

Support V is for Voltage