Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Burton]

Pressure Transducer woes ...

Yesterday while riding home I rolled off my throttle to have my bike starting to regen ... it is programmed not to as I have a footbrake programmed clutch for regen control.

It did this for about 1/4th of my trip (or 1/2 of my highway trip) and then it didn't do it anymore. I thought it would be a loose wire, again, but it returned to normal so I thought I would have to check if it is really regen or just wind resistance today.

Hop on the bike this morning,and find the bike has 100% regen (which is a lot since I have reprogrammed it) when I roll off the throttle ... it is 35F outside.

Roll bike into garage and look at voltage output of the transducer and it is 4.00 ... I squeeze my clutch and it reads 3.89 ... great, a bad transducer I thought.

So I proceed to try and track down my baseline DCF file and cant find it anywhere. Thankfully someone provided me with one and I normalized it between my current save to get a DCF I can roll back to when the transducer breaks (this is my third one mind you)

I get the DCF uploaded and look at the input screen to make sure things are set right to find the transducer at .43vdc >__< ... so my mind is thinking ... ok it really was a loose wire. I roll back yet again to my previous "current" build with the footbrake activated and try to wiggle every wire in the transducer circuit to see if it jumps back up and sticks at 4.0vdc

Nothing I do can reproduce it so I am beginning to think something, be it the transducer, the controller, the dc:dc, is getting cold and messing everything up. My money is on the cheap transducer even though it is filled with oil and not something like water.

Anyone have experience with transducers being fickle in near freezing temperatures??

[panurge]

Hey Burton

I don't know what kind oftransducer you are actually using, but yes they could suffer from freezing temps, depending on the type and specs of the device. but as you point out these are mostly problems concerning the water ice expansion. I belive that, likely, even in anoil based system, freezing moisture even in the enclosed air could lead to problems with some transducers at freezing temps.

Since me too, I'm working to develop a left (hydraulic clutch) lever for regen, I asked here and there my expert buddies about the best reliable and precision hydraulic device for this application and got the advice to try an industrial sealed pressure transmitter or a crystal type transducer, often devices rated for extremes temp ranges, able to take vibrations and last longer cycles than standard devices, althoug more expensive. Have not really deep-searched about them actually, but just typing on google gives good examples with specs and everything.
It is a very interesting application, and will follow your reports and develops...


A bit OT....^_^

My ICE biased idea for a regen actuator is to work from a clutch lever, maybe an hydro kit like brembo and other makes, even with a racing foldable lever (like these ASVs http://www.htmracing.it/cms/images/foto_prodotti/foto_frizioni/frizioni11.jpg). what would be great, is to use then a sort of multiswitch pressure device allowing the lever to act with bands like:
- from 100% of throw (totally squeezed) to say 85% = 0 regen
- from 85% to 50% of throw = 40% regen
- from 50% to say 15% = full regen
- from 15% of throw to release position =  normal regen (that means the roll back regen settings of your actual mode)

This way the bike could still works exactly in the "factory" way, if you're not touching the lever (and even fold it up), but can give you a -0 regen coast feeling- while squeezing the clutch and a -clutch plates engage feeling- while releasing the lever, till the normal rollback setting you have....

A more electric biased approach could be otherwise a Tesla style twist throttle, that in a motorcycle should give something similar to the engine braking feeling of Big mono or twin ICEs bikes while releasing the throttle partly/fully with slow or fast movements.

J.L.

[Richard230]

My 2014 Zero S is completely stock, but sometimes I find that when the humidity is around 100% my "custom" programming will disappear and the bike will revert to "sport" programming.  After riding for a while and things dry out a bit, my "custom" programming will reappear.  Kind of odd.   

[Burton]

I have been thinking and figured I would just let some of those thoughts out here.

Options for regen:
1) Mechanical Clutch with potentiometer hopefully in a water sealed box.
2) Hydraulic clutch with pressure transducer
3) Another "throttle" on left handlebar (uses potentiometer)
4) Bike clutch with pressure switch
5) Vetrix style (which is a PWM output and wont work stock with the sevcon gen4)

After riding this way for a while I am not sure if having a "second" throttle would be easy to use. Even if you capped the voltage range I think when you start to weight your palms from braking forces it might be difficult to adjust your input at the level you can with a clutch.

This might also be said for the vetrix style input ... but since I don't have a vetrix I can't really compare this. However since I haven't seen many complaints about the throttle being hard to adjust for regen upon braking hard I will assume it is a valid option.

So ... this brings us to the issue at hand. My transducer has something wrong with it and I can't tell you what it is as it is hard to isolate on the spot but I am leaning toward freezing issues.

My potentiometer uses the CAN 24vdc out from the controller and runs it through a drop down dc : dc to 5v then to the transducer then to the analog 2 input on the controller. (You can't use a power source outside the controller as it won't work)

That is a lot of places something can go wrong and it is intermittent and often only happens on the road. To combat this I have downloaded a stock 2013 S dcf to my sevcon controller then commissioned the motor and saved it again. This is if something happens to my transducer I can swiftly change the programming out so I can safely get home.

So I think temporarily trying dual throttles might be smarter than trying to find a good quality (read $300) transducer which is often much larger than the one I have to "fix" this problem. A new throttle would cost $100 shipped and would be easier to set up since it can accept higher voltages. But it will still have to use the CAN PWR 24v supply and step it down to 12v or add a resistor inline

There is something else I am working on but that solution won't be viable for a while


@Richard
Yes yes yes ... this has happened to me before as well. I find simply restarting it helps with a small wait between restart. No clue why it does this but it does and it is far harsher when your torque is set at 39Nm vs the stock 12Nm when that 7% "stock" regen setting kicks in

[panurge]

After riding this way for a while I am not sure if having a "second" throttle would be easy to use. Even if you capped the voltage range I think when you start to weight your palms from braking forces it might be difficult to adjust your input at the level you can with a clutch.

This might also be said for the vetrix style input ... but since I don't have a vetrix I can't really compare this. However since I haven't seen many complaints about the throttle being hard to adjust for regen upon braking hard I will assume it is a valid option.

Although the patented PWM neg/zero/positive throttle works great and intuitively,  issues with the front brake lever are not uncommon with Vectrix and negative throttling; my buddy who owns one,  has the brake levers swapped, for that exact reason.
 I wouldn't have any negative throttle, nor a second left throttle in a true motorbike...I mean...these could be valid and perfectly working options for some, but that's not my personal feeling.

The hydro clutch concept is my preferred one, but as stated in my last post I would try to almost reverse the concept  with 0 regen at full lever's squeeze, and, through customized regen bands, still maintaining the factory settings while not pulling the lever.

I'm pretty sure, that kind of pseudo varible clutch regen should be achieved even without mods to the controller, and only working at the 12v power line level and with actual buttons and sensors.
I'm sure that using a small, external microcontroller, shouldn't be so hard to get that kind of system only acting with the mode switch and the brake lights......indeed, actually, there are 6 different regen settings in a bike (2 for each of the 3 settings), but you can't freely cycle among them, a microcontroller with some code should do that easy and in a very small box......

[Burton]

The hydro clutch concept is my preferred one, but as stated in my last post I would try to almost reverse the concept  with 0 regen at full lever's squeeze, and, through customized regen bands, still maintaining the factory settings while not pulling the lever.

This is just a matter of switching high low (start stop) variables in the DVT setting.

I'm pretty sure, that kind of pseudo varible clutch regen should be achieved even without mods to the controller, and only working at the 12v power line level and with actual buttons and sensors.
I'm sure that using a small, external microcontroller, shouldn't be so hard to get that kind of system only acting with the mode switch and the brake lights......indeed, actually, there are 6 different regen settings in a bike (2 for each of the 3 settings), but you can't freely cycle among them, a microcontroller with some code should do that easy and in a very small box......

I am not sure if you can change the "coasting" vs "regen" behavior on the fly via the CAN or not ... but this is what you are describe. That is no variable "engine braking" controlled by a clutch but only a switch to change between coasting and regen on braking.

The settings you would have to change on the fly via the ODBII would be

zeroneutral     - Neutral When Off Throttle      :  No / Yes
brakeregen      - Apply Regen On Brake        :  No / Yes (if desired)

But again .. this would be more like a switch than a clutch And I don't know what effect there would be if you did this on the fly (like having to reset the throttle to 0rpm before it took effect etc) And you wouldn't need a power source from the controller since the changes will happen via the ODBII

fully variable regen is really important to the way I ride. The hyrdo clutch is nice because it is like a pressure switch the way it feels when in use but I might go purely mechanical clutch with a pot box just for reliability.

[panurge]

Mine was just a concept idea, exactly as you describe, not a true varible regen but a switch (I've used the word "multiswitch" before), I've called it "pseudo-variable" for that reason. Eheh! obviously a true variable regen as you use and suggest is way better, but not for everybody......if I will own a Zero, sure I would go with the full variable regen even if that would mean to virtually loose the warranty, but, actually, I have a friend/customer owning a DS and asking for a way to get some extra regen features without the need (and the implicit risk of warranty expiration) to reprogram anything.

Good point about the need for 0 throttle before changes take effect....my feel is that it is true for speed settings but not for regen settings where changes seems to be valid just after a fraction of a second, at least in the 2014DS I've tested recently for some days, but maybe I'm wrong.
Were this the case, pulling/releasing fast the clutch lever, one could luckily pass through all the lever's bands of my switch idea under the limit time, and go directly to the other extreme (from zero neutral to the actual throttle roll back regen and vice-versa) but even, with a bit of use, pull (or release) the lever slowly (band by band) and give it the time to get the relative regen.
There's even to consider that one would pull the regen lever (and get regen or a coasting feeling) mostly while rolling back to 0 throttle or yet there.

I would just like to try that, and maybe the real world would say that's totally wrong and useless.......

A left hydro brake lever plus pressure transducer is probably the best performing and elegant solution for fully variable regen, indeed.

[mrwilsn]

Options for regen:
1) Mechanical Clutch with potentiometer hopefully in a water sealed box.
2) Hydraulic clutch with pressure transducer
3) Another "throttle" on left handlebar (uses potentiometer)
4) Bike clutch with pressure switch
5) Vetrix style (which is a PWM output and wont work stock with the sevcon gen4)

Are you familiar with these strain gauges? I haven't had time to do any in depth research so I only know what's in the video but it seems like it would work with a clutch/brake lever. Then you wouldn't need any fluids but you might need to increase the resistance to pulling the lever so that there would be enough flex in the lever for the strain gauge to detect (or maybe not...I'm not sure exactly how sensitive they are)

https://www.youtube.com/watch?feature=player_detailpage&v=IxB2j-egWcQ#t=2010

Or maybe one of these? I don't think Grin Technologies is making the universal throttle yet but you might be able to get one of the force sensitive resistors and create your own version.

https://www.youtube.com/watch?feature=player_detailpage&v=IxB2j-egWcQ#t=5480

Sent from my XT1060 using Tapatalk

[Burton]

Yes Mrwilsn, in fact #4 I meant to put strain gauge but instead put pressure switch >__<

The strain gauge has been done before with a bicycle hand brake on a FX without the need of messy brake fulid

[Burton]

A left hydro brake lever plus pressure transducer is probably the best performing and elegant solution for fully variable regen, indeed.

Followed closely if not paired with the mechanical potentiometer in a box attached to clutch levers

[panurge]

A left hydro brake lever plus pressure transducer is probably the best performing and elegant solution for fully variable regen, indeed.

Followed closely if not paired with the mechanical potentiometer in a box attached to clutch levers

Yes, of course, and maybe more reliable but less elegant.
I have a nice box made by an endless-sphere member that's for mechanical wire to hall effect, and a lot of potentionemetrs for the Oxygen cargo scooter that are leverage type pot, with an insert on the leverage to hook a standard mechanical wire head onto, I'll post some specs/pics when at home.....it is reliable and interesting, made in UK.

[Burton]

Yes, of course, and maybe more reliable but less elegant.
I have a nice box made by an endless-sphere member that's for mechanical wire to hall effect, and a lot of potentionemetrs for the Oxygen cargo scooter that are leverage type pot, with an insert on the leverage to hook a standard mechanical wire head onto, I'll post some specs/pics when at home.....it is reliable and interesting, made in UK.

You mean this one ? https://endless-sphere.com/forums/viewtopic.php?f=31&t=58801 5vdc if I recall.

[RNM]

We've talked about strain gauges before, here: http://electricmotorcycleforum.com/boards/index.php?topic=4327.msg27650#msg27650

[panurge]

Yes, of course, and maybe more reliable but less elegant.
I have a nice box made by an endless-sphere member that's for mechanical wire to hall effect, and a lot of potentionemetrs for the Oxygen cargo scooter that are leverage type pot, with an insert on the leverage to hook a standard mechanical wire head onto, I'll post some specs/pics when at home.....it is reliable and interesting, made in UK.

You mean this one ? https://endless-sphere.com/forums/viewtopic.php?f=31&t=58801 5vdc if I recall.

Yes exactly.
The max output is 6VDC and it is adjustable by moving the magnets inside the box.
The Oxygen Cargo Potentiometer, should be the best option for a cable actuating pot, though. It is splashproof, reliable and made for Heavy Duty industrial applications. Shouldn't remember the brand and shouldn't find the specs on my laptop right now, It is certainly in my Shop deskPC and anyway I have them phisically so I'll post an image soon, there are versions with almost any output.....

[panurge]

as promised, that's the oxygen cargo's potentiometer: