Doubt in pneumatic system

[Thrilok]

Hello everyone
For the season 2011 our team is making efforts to implement an electro-pneumatic gear shifting system.

we have planned a preliminary basic circuit in which we are planning to use paint-ball tank for air storage and a double acting pneumatic cylinder(actuator) for shifting purpose(having the initial position of the piston rod in the middle of the cylinder) on which three proximity sensors will be placed to position the piston.

we have a 42.5mm lever on the gear shifting spindle, for which approximately 10.5mm stroke on either side is necessary for each gearshift.

we measured the stoke length manually and not sure of it, so we are afraid if the actual stroke is 10.4mm or 10.3mm and we calibrate the pneumatic cylinder to have a stroke of 10.5 mm, during each shift the the pneumatic actuator will exert a force on the lever even after it has reached its extreme position which will damage the return-spring and other parts too, eventually we may end up with broken parts.

so I request if anyone can help us by guiding in calibrating the exact stroke for each shift or suggest any other alternative methods to avoid extra force on the gear lever after the gear shift.

[RenM]

Hi Thrilok,

We never had any issues damaging the shifting mechanism and we dont use anything to limit the force at the upper or lower position of the piston.

[Some Guy]

Take a look at the stock shifter setup from whatever your engine came out of, in our teams case a sport bike. There is a lot of force multiplication between the riders foot and the shifter. Now consider how the average sportbike rider treats there equipment and how long it actually lasts under such duty. You may end up doing what RenM does.

[Nicky]

We used proximity sensors before but then shifted to shifting based on pulse widths to avoid extra sensors and wiring. We did on a trial basis. Start with the a pulse width of say 150ms and work your way down till it stops shifting. A general rule of the thumb is to add 20ms to this and let it shift. We even tried tweaking the pressure at the regulator to get more shifts.

[Dash]

Use a pressure regulator so that you can't put 10,000 psi on the arm?

[Thrainer]

Originally posted by Nicky:
We used proximity sensors before but then shifted to shifting based on pulse widths to avoid extra sensors and wiring. We did on a trial basis. Start with the a pulse width of say 150ms and work your way down till it stops shifting. A general rule of the thumb is to add 20ms to this and let it shift.

This is pretty much what we did, but with a double-acting solenoid. For two years, we had highly sophisticated electronics and sensors, which didn't really work as planned. But the solution which is a lot less effort and much more reliable is a pair of automotive relays and some programming in the M400 (used successfully in 2009).

Regards,
Thomas

[Garlic]

Size your pneumatic cylinder / pressure combination so that it does not exert a force that can damage the engine.

Size it so it has enough force to reliably shift, and nothing more.

Problem solved.

[Thrainer]

Garlic, how exactly do you suggest to find that number? A destructive test with the transmission?

I don't believe there's a certain force "enough to shift", because this number depends on several factors. The faster you want to shift, the more force you need. Further, it depends on how your transmission is loaded (up or down shifting, throttle position and RPM).

[Shields]

Dependent on how complex you want to get with your programming, you can alter the shifting parameters based on input (for example, running a different time or pulse for downshift as opposed to upshift..)

In my experience with sportbikes, these transmissions can take alot of abuse as long as you dont play with changing the internals (like shift drum)

Also you want to make sure you are not destroying the actual gear shift lever etc and those are easy enough calculations to make based on input force etc.

Not sure what solenoids you are running but many of the micro solenoid valves have a maximum pressure of 150psi.

[nms553]

I have been working on an electro-shift system for the same reasons. however, with the pneumatic systems the last couple years we have always had trouble with bending shift forks (yamaha YZF R6). I believe this was happening because the actuation time and shift force weren't tuned, forcing the shift forks to "wall up" against the gears before the gears could disengage. i propose live testing with a conservative shift force and actuation time and ramping them up from there. However, a greater force ensure shift reliability, so introducing an energy absorbing instrument (perhaps a stiff spring) for those shifts that "wall out" may help you achieve that reliability without paying the consequences. also, depending on what motor you have, I don't think the r6 motor is as "plug and play" as the cbr.

[Nicky]

We are running our pneumatics through the M800. Our onboard compressor charges upto 10bar. It's very slow and annoyingly loud. Do teams run onboard compressors? Can I make do with an off board one and charge a canister?

I was thinking of a paintball canister. But don't have a compressor for that pressures. Would a 1ltr 20bar charge be enough for endurance?

I tried searching for a better compressor but couldn't find one. Please suggest.

[Gaanja]

@ Thrilok: A difference of 0.1 mm on the stroke length will not have any damaging effect. These engine are built with a good factor of safety and to take a lot of abuse.

[Canuck Racing]

Nicky, do you have any pictures of your compressor on the car? That might help answer your question.

[Shields]

Alot of teams I've seen have used paintball tanks. Either CO2 or air/nitrogen/argon etc.

You should be able to rent a large fill tank and buy a remote fill station to be able to fill the smaller tank. Last year we rented a SCUBA Tank to fill our compressed air tank for the pneumatics. You just need to ensure that you properly size your tank to be able to last endurance.

As for "should this be enough for endurance" --> do the math. You should know bore, stroke, pressure and air specs, thus being able to find out mass/volume of air used per shift. Rough idea of shifts/lap*22 laps = roughly what you need for endurance + a healthy factor of safety.

[DanielH]

We use a paintball tank filled with CO2 for our shifter and clutch. (the clutch proved to be a hassle so we are moving back to manual actuation). But it caused us to have a DNF at california last year. because we ran out and the connection was loose. Our big CO2 tank we use to recharge the onboard tank wasn't pressurised enough to fill or onboard tank. we also had problems with valves freezing do to over use of the shifters.

we are thinking of switching to nitrogen/air next year because its more stable and you can get more into the same sized bottle.

[Nicky]

@Canuck Racing: Here's the picture of our onboard compressor. It's about 800gms and takes around 4amps when charging at 7bar.
http://www.hardracing.com/images/compressor1.jpg

Regards,
Nicky

[Thrilok]

@Nicky compressors that can compress air upto 200 bar or so come very costly, instead you can fill your paintball tank using a scuba tank which you may get easily around your place.

[carditommaso]

Hi guys,
I know that the best and simple way is to rent a scuba bottle,
But, if you are think like selling your car, is not very simple explain at the buyer : " you have to go almost everyday to fill your scuba bottle!"
So please, if someone used the nicky method, tell us about volume of the bottle and eventually battery problem.

[Lorenzo Pessa]

Originally posted by carditommaso:
If you are think like selling your car, is not very simple explain at the buyer : " you have to go almost everyday to fill your scuba bottle!"

there are many scubas don't think this is a problem :-)

4 ampere are a lot. You loose some HP when the pump is working.
Which is the weight of the overall system?
Have you evaluated if a manual shifting would be competitive compared to your system?

About pressure, our system work at 200 bar (on an Aprilia 550). we get to the competition two scuba tanks. the first one to refill for testing, skidpad, acceleration and autocross. the second only for endurance.

[Nicky]

@Lorenzo: we got a lighter .7ltr canister which saved almost 1kg, we did away with the mounts that came with the compressor and manufactured our own, which further saved weight.

The entire system now weighs 1.2kgs. The advantages over the manual shifter is giving the driver more time on other things while changing the gear is only a button-press away. also we have ignition kill and hence, no more throttle blipping or clutching for upshifts. all of this has been tested. we burnt a set of clutch plates, but quickly adapted. we now have a stable ignition kill/retard values programmed in. Shifts are now 90ms including kill time and the time required to bring power back to 100%(we slowly get back the power on the engine thro fuel and ign so as to not overload the clutch).

thats it for now. may encorporate an auto downshift feature based on tps, brake and rpm, but would reccomend against it.

As for HP decrease while the compressor running, each shift require 3sec of compressor running to recharge the air utilised.
From my calculations, 8-10shifts/lap of endurance=24sec of charging time. Something that we could afford to lose an HP over.

The reason that we have all of this is to avoid cost in refilling and the cost of plumbing for high pressure lines. Not to mention the cost of the canister. Hope that helps.

[Lorenzo Pessa]

I like this kind of complete answer! Thank you.

See you in Varano.

[Nicky]

Looking forward to meeting you in Varano, too.

[Nicky]

A small update for you guys on this. Our compressor gave up on the 22lap of endurance. Caused a DNF. I was so sure of this as i'd seen em being used on sports bikes all over. Hard luck. Prolly could have been with running it at higher duty cycles or not providing enough cooling.