I've cured my Vectrix - It no longer tells lies

18 replies [Last post]
Offline
Joined: 07/30/2009
Posts:
Points: 248

Hello Folks,

Since the day I bought it, my Vectrix bike has told lies. Sometimes, not a problem, other times, well, BIG problem. Have you ever tried pushing a 'dead' Vectrix? Yes, I know, stupid question. Of course you've had to push the Bloody thing and usually, because it's been lying again.

The 'lies' were one of the problems. Besides the battery failures, the lies were probably the next worse thing about the bike. Mine would often suggest that I still had 6 or 8 miles to go, would show perhaps 3 to 5 bars on the 'fuel' gauge and then a hundred yards (metres if you're metric) down the road it would crawl to a virtual halt and I would know that it had been lying again.

As you may know, I have been 'modifying' the vectrix charger behaviour. Initially, after re-arranging the charger in order to prevent 'heat damage' to the battery, I dumped the useless fuel gauge and estimated mileage readout in favour of using the 'left brake/kill switch' combination to allow me to estimate the miles to go. For some ten months that system has worked for me, and perhaps for one or two of you who have been brave enough to try my modified programme.

Just a few weeks ago, I was asked if it might be possible to re-jig the programme to work with the older Motor Controller which, I was told, gives a much more lively performance with it's better acceleration. Well, I never believe everything I am told, so I tried the earlier Motor controller and, yes, it does give a more lively ride and, yes, it does have better acceleration. My interest was captured with a new challenge. It seemed time to 'try again'.

I have learned much whilst working on the programming and felt more able to get the results required. Those results being. An accurate working fuel gauge. A more accurate estimated mileage readout. A charger operation that does not damage the battery. More even control of the battery temperature and one or two more minor improvements.

I have basically succeeded. It has been a long and tedious job, but I now have arranged twenty five changes in the programme in order to achieve the above improvements.

So what is the result of all of this?

Well, the end result (which is all that really matters) is that I have a charger file which, in conjunction with he 1014b Motor Controller, :-

1/ Charges the battery up to over 95% of the actual capacity (up to 30 ampere hour) of the battery without the normally associated heat damage.

2/ Fills up the fuel gauge according to the ampere hours put (charged) into the battery. (this means that 'full' / 17 bars represents 30 ampere hr) If the battery will not accept 30 ampere hrs, say it only accepts 24 ampere hrs, then the fuel gauge will reach only about 13/14 bars. This gives a direct indication of the battery capacity as it is, not as you might wish it to be.

3/ The estimated miles display now reads more accurately with faster updates. Maximum estimated miles is around 68, with a full fuel gauge (17 bars) proportionately less with less bars. On regenerate the display will show a maximum of 75 miles (a default setting) again this is with a full fuel gauge.

4/ The fans run on for ten minutes after switch off, and also if charging is terminated by unplugging. This is a fixed setting.

5/ The fans also run whilst riding and whilst charging.

I have also 'raised' the theoretical 'battery losses' to be more in keeping with the actual self discharge of the battery.

To date, I have not had the bars suddenly disappear with a corresponding loss of battery power. What happens now is that the bars run down and the estimated miles reduce as the battery charge is depleted. Near the end of charge the last bar will 'drop off', the estimated miles will be reading 1 or 0 and there will still be enough charge to run a little further ( mile? maybe a bit more). I have not had a power loss and still had bars displayed.

No more lies. No more 'range anxiety'. I now know when it is going to stop, well in advance and fairly accurately.

My own bike which was charging only to around 20 ampere hour a few weeks ago, (I knew that I had a battery problem and this was expected), has just taken a 27 amp hours charge. ( I now get well over 30 miles on a charge and this is in hilly country.) My 'damaged' battery is actually improving.

Another bike, which I have installed this programme into, had been standing for some months following an accident, the repair and the owners reluctance to use it for travel to work due to 'range anxiety'. This battery is now showing 23 ampere hour capacity, although when it arrived here two weeks ago, it showed less than 15 ampere hour. It has now been charged five times and is showing a 8 ampere hour improvement.

Just to restate what I am using:
I am using the 1014b Motor controller software with my modified charger programme.
There is no 'left brake /kill switch' facility (for reading the voltage and temperature – that comes only with the later motor controller software).
There are no more 'lies'.
There is no more 'range anxiety'.
There is no more 'pushing a dead Vectrix'.
The actual battery capacity can now be 'seen'
on a charge from empty to full. If, at the end of the charge, there are fifteen 'full bars' then the battery capacity is (approximately) fifteen seventeenths of thirty ampere hours i.e. twenty six ampere hour.
There is no excessive temperature rise when charging the battery (not more than 3 degrees C) in fact the temperature sometimes drops whilst charging.
There is no need to 'cool' the battery before charging as the battery no longer gets overheated.

Whether the above is an improvement or not rather depends upon your point of view.

I am looking forward to Xvectrix's point of view on this one and I shall also look forward to your point of view as a Vectrix owner/rider.

One last point. I am willing to share my achievements with you all. I am offering this information and files for your interest only. I am not suggesting that you install these files or use the information in any specific or non specific way. You must accept total responsibility for any use that you put this information and/or files to.

Telling it like it is, as always.

The Laird.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Domi422's picture
Offline
Joined: 03/21/2009
Posts:
Points: 50
Re: I've cured my Vectrix - It no longer tells lies

Thank you very much for the huge work you have done and for offering to share.
My vectrix is now out of warranty and with all the improvements you are announcing it makes sense for me to buy the canbus adapter and try your modified programs.
I've no knowledge in elctronic but know how to use my mind and hands so I think I could go through the update procedure without trouble.
I will let you know when I get my canbus.

__________________

Le site des utilisateurs français (The French user's website) http://www.vectrix.fr/

Offline
Joined: 10/20/2008
Posts:
Points: 369
Re: I've cured my Vectrix - It no longer tells lies

All i can say is thankyou Mr Laird.....
I was a guinea pig for his software and i am very very happy with it. Unfortunately i don't have access to the canbus adaptor presently so i have a slightly older 'pre-production' version of the software described above.

I have to say that everything he says about the cooling etc is 100% accurate. After a 90kph 15km run home on my bike (with ambient temps around 20 degrees presently my batteries will be around 25-27degrees when i reach my destination. Whilst charging this drops to maybe 22/23 degrees throughout ! Obviously i am not in the sort of silly temps Northern Australia gets so i can't comment on that.

The only thing that Mr Laird didn't say was that in order to keep the temps down (i believe) it takes longer to charge the bike. however on the upside of this is that if you are like me and you tend to plug in after a commute home and leave it for the timer plug to activate, a minor change of timing is all you need to do.

Oh and in reference to the reason for using the 1014b software. Basically is this, i've been running my bike on that version of software as my temp sensors failed and the system fails to charge with faulty temp sensors using the MC1017 (latest version) of firmware as it checks the temps and then shuts off the charger immediately. So i didnt know what the 1017 and the 1017 (68mph - 115kmh version was like) So i was happily running my bike on the older firmware. I replaced my temp sensors (thanks antiscab) and proceeded to install 1017-68mph. I rode it for ummm 2 days like that. Oh my god....it turned my bike from something ridable to a complete SLUG in comparison. Absolutely dull as dishwater and seeming complete inability to accelerate faster than a diesel bus (seemingly). I would estimate possibly 20-25% LESS torque than the older software. I mentioned this to mr laird who tried it and was surprised at the results. Yes as you know acceleration means current which means temp and also less distance and yes i ride like a nana most of the time to give me the best possible range on the lowest amount of bars. However there are quite simply times when i want/need to get out of a situation or blow out the cobwebs and i want some acceleration then. This version of the motorcontroller will give you that. Oh obviously its limited to 105kmh or around that figure unlike 1017 version. However i reasoned that the 1017 verison takes so long to get to 115kmh and 90% of my riding is 40-80kmh that it makes it pointless to have anyhow. Try it for rideability.

So.... in conclusion. Mr Laird you are a f*****ing genius. Welldone, quite frankly if you'd written this a few years back there'd be less batteries dead i'm sure. I am massively in debt to you (and antiscab for parts- and michaet ) thankyou for making my NiMh version the best it can be without me damaging it .

People if you have a canbus adaptor - try it, i doubt you'd go back to the vectrix or 1017 versions. Mine has been running it for ummm probably 6-8weeks daily. Awesome absolutely awesome.

jdh2550_1's picture
Offline
Joined: 07/17/2007
Posts:
Points: 2338
Re: I've cured my Vectrix - It no longer tells lies

Congratulations! That's no easy task and it sounds like you've done a great job.

I'm curious as to how you made the modifications? (No need to say if you don't want to - but I am curious)

__________________

John H. Founder of Current Motor Company - opinions on this site belong to me; not to my employer
Remember: " 'lectric for local. diesel for distance" - JTH, Amp Bros || "No Gas. No Worries." - JDH, CuMoCo || "Make Volts Not War" - anon.

Mik
Mik's picture
Offline
Joined: 12/11/2007
Posts:
Points: 3710
Re: I've cured my Vectrix - It no longer tells lies

The Laird wrote:

...
... Initially, after re-arranging the charger in order to prevent 'heat damage' to the battery, I dumped the useless fuel gauge and estimated mileage readout in favour of using the 'left brake/kill switch' combination to allow me to estimate the miles to go. For some ten months that system has worked for me, and perhaps for one or two of you who have been brave enough to try my modified programme.

Just a few weeks ago, I was asked if it might be possible to re-jig the programme to work with the older Motor Controller which, I was told, gives a much more lively performance with it's better acceleration. Well, I never believe everything I am told, so I tried the earlier Motor controller and, yes, it does give a more lively ride and, yes, it does have better acceleration. My interest was captured with a new challenge. It seemed time to 'try again'.

I have learned much whilst working on the programming and felt more able to get the results required. Those results being. An accurate working fuel gauge. A more accurate estimated mileage readout. A charger operation that does not damage the battery. More even control of the battery temperature and one or two more minor improvements.

Fabulous! I've tried the latest official version for a while now, with battery capacity set to 18Ah (IIRC - it might be 16Ah). But, due to the usually incorrect estimate of self-discharge, the "lying" continues. Little lies, mostly, so I never had to push recently, but dangerously low power and reversing cells have occurred. And of course, the pack gets hotter as Summer is approaching. I've tried your 864W charger version before, it worked very well, but the gauge was not yet functional. Do you still use this single stage power setting for charging - or something more complex?
How does your charging algorithm determine the (95% of the) batteries actual capacity? Will it be able to cope with my capacity-imbalanced pack, limited to about 16Ah by the half-dozen weakest cells?

Quote:

I have basically succeeded. It has been a long and tedious job, but I now have arranged twenty five changes in the programme in order to achieve the above improvements.

Thank you for all these efforts and for sharing the results - much appreciated!

How much does your improved charging algorithm depend on properly functioning temperature sensors in the battery pack? Mine seem to have a problem, the max temp reported is always about 6degC too high. Recently, battery symbol and sometimes also the temp warning are flashing, but the impellers do not start up reliably when a pre-charge cooling period is selected. With battery temp almost at 40degC (reported as 45degC), and clearly above ambient, the impellers sometimes stay off when the charge delay is running.

I'm keen to try this out!

Cheers, Mik

__________________

This information may be used entirely at your own risk.

There is always a way if there is no other way!

Offline
Joined: 07/30/2009
Posts:
Points: 248
Re: I've cured my Vectrix - It no longer tells lies

Hi Mik, Hi Everyone,

It looks like my post has produced a lively response. I shall try to answer all questions as they come up.

I am using constant current in two charging stages/steps. First, the battery is charged at 6amps until a preset voltage is reached, this is followed by ten minute 'breather' and the a 2 amp constant current charge follows to fill up the battery, again to a preset voltage. The preset voltages I am using have been determined by experimentation and have proved to be correct for a battery in good or reasonable condition.

Good or Reasonable condition. What that means is that the battery voltage, regardless of capacity, can safely reach the preset voltages whilst being charged. Missing or short circuit cells will affect this process. However, it is not too difficult for me to change those preset voltages for specific situations, I am open to requests on that one.

The 95% charge is determined by the batteries capacity not by the charger programme. No matter what he capacity of a battery, the voltage will rise when the battery is full and this is what is being used to end the charge.

Where cells of different capacity are in series connection, as in the Vectrix, the low capacity cells will determine the point at which the battery fails to supply useful energy on discharge. This occurs when the low capacity cells are depleted, the good cells will still contain considerable charge. Therefore the low capacity cells will charge up along with the good cells and the use of a preset voltage limit for charge determination should work satisfactorily.

Just recently (about ten/twelve weeks ago), I all but vandalised my own battery. I was returning home after a 35 mile ride and the journey was up hill. With about two miles to go, the bike began to lose power, with one mile to go it all but stopped, it was capable of only walking pace. It stopped completely with about three quarters of a mile to the top of the hill. I waited a few minutes and it had recovered enough energy to move another 100 yds. Again I waited and again it recovered enough energy to move another 100yds. I have serious difficulty in walking due to an illness, pushing the bike, even on the flat, is all but impossible for me and i was determined that this bloody bike was going to take me home. It was over an hour later that I gained the top of the hill and it was now down hill all the way home.

The result of that Vandalous treatment of my bike did the battery a serious mischief. The next charge gave me only about 18 miles and I knew then that I had battery damage.

Since the charger programme re-write, That new programme has been in my bike for about three weeks. I have done around 250 miles with maybe ten/twelve charge cycles and the last mileage was 34.6 on a full charge. When the programme first went in the bike, the fuel gauge only achieved eight bars. On the last charge the fuel gauge has filled up all seventeen bars and I shall be going out later to see how many miles it can do this time

In all of this, the battery has not 'heated' whilst being charged and both the programme and the battery seem to be working most satisfactorily. Time will tell as they say.

The temperature sensors are largely irrelevant as the temperature now does not rise more than a few degrees over ambient. The temperature limitation within the programme is designed only to shut down the charger in the event of a serious overheating problem. I have never tested it's operation and there are other safeguards which should always operate first. The programme will work without the temperature sensors being present or so I am told.

As for pre-charge cooling, The modified charger programme usually results in the battery temperature dropping whilst on charge. Obviously this will depend on the battery temperature as related to the ambient temperature but it does occur regularly in my own experience. Ambient here is at 10 degrees C outdoors. In my garage where the bike is charged, it is about 13degrees C. Bike battery temperature ranges from 14 / 17 degrees C in use and on charge. Remember that my battery is fitted with equalising resistors which keep the battery at three degrees above its normal temperatures AND i have also fitted two low power (6 watts each) fans in the plenum chamber. So draw your own conclusions on the heating effects.

John/JDH2550_1, send me a personal message with your questions and I will provide the answers which are not easily dealt with here.

Mik, send me an e-mail stating exactly what you need. I'll get to it as soon as I can.

XVectrix, Where the hell are you? We have things to discuss and I'm still awaiting your reply to endless personal messages and requests for contact via this forum. Have you got a problem? I'd just love to hear from you.

Keep smiling folks,

The Laird.

(Telling it like it is, as always).

moccasin's picture
Offline
Joined: 07/11/2008
Posts:
Points: 494
Re: I've cured my Vectrix - It no longer tells lies

Congratulations on th excellent results from all your hard work!! Sounds like you nailed what Vectrix should have done in the first place, by simply rethinking the process.

My bike has the newer software installed, and you are right, it kills the "performance" of the bike in favor of adding a few more speed notches on the clock. In my own world, my vectrix never goes anywhwere that the speed limit is over 50mph, so the older 62mph was more than adequate, but with the new software, anything at or below half battery guage results in less than 50mph top speed anyway.

My daily use of the Vectrix is mostly less than 15 miles, and more often than not, less than 10, so she tends to "range down" after a couple of week's use, often leaving me at walking speed having only gone 17 miles. A more acurate battery guage would be a huge improvement.

I'll have to admit that I am totally not into the "techie" side of these bikes. I'm the poster child for "Consumer Test Dummy" as I ride the thing to and from work, and for errands during the day, then plug it into my timed outlet for nightly charging. I ride it, plug it, and forget it, and with the obvious limitations that has worked well for the last three years. So, I'm clueless as to how to put YOUR program into MY bike, and I wouldn't know a canbus from a greyhound bus, so I'm out of the ballkpark for this wonderful acheivement, but will watch with great interest and wish you all the best, and thank you again for putting the V back on the map again!! :-)

Offline
Joined: 07/30/2009
Posts:
Points: 248
Re: I've cured my Vectrix - It no longer tells lies

Hello folks,

I'm back from that ride. Just did 39 miles on a full charge.

To be honest, it did start to lose power on the last two bars although it did keep moving 'till they were all gone.

Perhaps the terrain I ride in is not to helpful here, lots of hill work and quite steep too.

Here is a chart of today's ride, just for your interest.

Route01.jpg

Anyway, despite that lack of power on the last few miles, I think that it did O.K.

Must go lots to do.

Still haven't heard from XVectrix, hasn't even been decent enough to tell me what his problem might be. Such is life and such are some people.

Have Fun Folks,

The Laird.

(Telling it like it is, as always)

R
R's picture
Offline
Joined: 06/18/2009
Posts:
Points: 1354
Re: I've cured my Vectrix - It no longer tells lies

39 miles is a good distance!

Offline
Joined: 10/08/2008
Posts:
Points: 341
Re: I've cured my Vectrix - It no longer tells lies

Wow, starting this 1000 foot climb at 30 miles is pretty brave! That is a great range given the elevation.

R
R's picture
Offline
Joined: 06/18/2009
Posts:
Points: 1354
Re: I've cured my Vectrix - It no longer tells lies

The laird, this firmware is really impressive. Since I started using it (v1 and v2), my battery capacity has increased form 24Ah to 29Ah. Now I can travel over 55 km again. And the charge always reaches 147v with very little temps rise. Congratulations!

RuFuS's picture
Offline
Joined: 08/13/2008
Posts:
Points: 69
Re: I've cured my Vectrix - It no longer tells lies

Great Job! Wonderfull...impressive work and i think all the community is glad of it!!!

So, can it be used in any vectrix... =D which are the main requirements? Would like to put it in mine!!!

Offline
Joined: 07/30/2009
Posts:
Points: 248
Re: I've cured my Vectrix - It no longer tells lies

Hi Rufus/Everyone,

This modified software is for use with the P1 charger only and is designed for use with the old motor controller software.
Which makes the answer to your question, YES as long as your bike has the P1 / ESD charger.

You will also need a computer, CanBus adapter and lead and a little technical ability.

Send me a personal message and I will do what I can to help.

Keep Smiling:-)

The Laird.

RuFuS's picture
Offline
Joined: 08/13/2008
Posts:
Points: 69
Re: I've cured my Vectrix - It no longer tells lies

The Laird wrote:

Hi Rufus/Everyone,

This modified software is for use with the P1 charger only and is designed for use with the old motor controller software.
Which makes the answer to your question, YES as long as your bike has the P1 / ESD charger.

You will also need a computer, CanBus adapter and lead and a little technical ability.

Send me a personal message and I will do what I can to help.

Keep Smiling:-)

The Laird.

Great...

Noob question:

- how can i check if my vectrix has P1/ESD Charger...?

Regards

Offline
Joined: 01/18/2009
Posts:
Points: 54
Re: I've cured my Vectrix - It no longer tells lies

I would love to sign up. I haven't messed with CAM in a long while but I'd love to upgrade my bike. Do you have a write up of what we need? I too want to verify that I have the correct charger, but I am pretty sure that I have the first Vectrix model. What is the best way to check?

-Cheers,
I

Offline
Joined: 07/30/2009
Posts:
Points: 248
Re: I've cured my Vectrix - It no longer tells lies

Hi Folks,

The best/easiest way to check the charger type is to look at the version of software which is installed. the P1 charger latest software is REV 3001. The earlier software(s) were REV 2008 and REV2012. Obviously you would need the use of the diagnostics programme and a CanBus adapter to read this detail.

The 'cheap' way is to take the front right (facing forward) side panel off the bike and look at the charger label. The label will have something like SW V2.008 and HW P1.48 : These represent S(oft)W(are) V(ersion)2.008 and H(ard)W(are) P1.48 :

Software numbers 2008, 2012, 3001 are, as far as I am aware, all software versions used in the P1 charger. I have no knowledge of the later (Runke/Chinesse?) charger. It might be useful if someone who has this knowledge were to post it for all to see?

There may be other ways, of checking the type of charger fitted, that I am not aware of. Perhaps someone will enlighten us all?

Have Fun,

The Laird

Offline
Joined: 01/18/2009
Posts:
Points: 54
Re: I've cured my Vectrix - It no longer tells lies

What CAN software do you use or recommend? I've done project communicating with the CAN Bus, but I haven't upgraded software (firmware) via CAN bus. I'm assuming we need to use the same software you use? If you don't mind educating me, how does a upgrade work? If it a compiled binary that upgrade the existing firmware? Or is the charger programmed via a streamed commands of the CAM Bus? I'm a developer myself, but pretty stupid when it comes to CAM. On my project where I worked with Ford we just listening for certain events on the CAM Bus... but most were from the CD player, or ignition. It was all for a Navigation System. It was always 'listening'. If you have a good link where I could read up on the topic that would be awesome so I don't waste your time. :D

-Cheers,
I

slbaker's picture
Offline
Joined: 11/22/2007
Posts:
Points: 59
Re: I've cured my Vectrix - It no longer tells lies

Sorry if this is a dumb question .. but where can I download the software and hex files for these modifications?

Thanks,
Scott

R
R's picture
Offline
Joined: 06/18/2009
Posts:
Points: 1354
Re: I've cured my Vectrix - It no longer tells lies

PMTTL

Private Message to The Laird...

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Short URL

JensonUSA Bicycle parts and accessories
Build Your Own Electric Vehicle You need this book if you want to build your own EV
Who Killed the Electric Car? Essential documentary explaining what happened the last time electric cars were killed
Customize This