Throttle by wire kits?
Discussion
Does anyone know if there are any aftermarket suppliers of the "drive-by-wire" throttle systems common to many new cars. Not looking for a specific model - bolt on application - kind of a custom application where I would build the adaptation of the servo motor to my linkage on the engine.
Well, I've totally rebuilt the induction system on my TR6 to have 6 sidedraft Keihin carburetors (motorcycle racing carbs) with custom built ram intakes which are all cable actuated. It runs extremely well, starts and idles like a kitten, sounds like a ferrari induction, frankly beats the heck out of the triple weber setup common to this car, but the problem is with my linkage which I keep breaking (the main cable pull to the foot pedal) due to the combined resistance of the carbs. In other words the problem is not at the immediate carb linkage but the difficulty of pulling strength via a single cable back through the firewall. Hence the interest in eliminating the cable to the pedal. The servo motor system seems a perfect fit for this application. And as you may have guessed, I am not completely sane, and love to mess with stuff on my TR6.
With drive by wire you would need an electric throttle body or some kind of motor to open the throttle. They are usually built as part of the ecu and the pedal sends signal to ecu then ecu controls the throttle opening. On an old car such as the TR it would be a mamouth task to try and set it up, if its possible at all.
Thanks for the comment. I would need a sensor in the pedal that activated a simple servo motor connected to my linkage - no interest in the throttle body. I know this isn't the common setup in factory drive-by-wire systems, but was just curious if an aftermarket supplier had something. I am in the US and there is a company called LOKAR that supplies a myriad of various throttle control assemblies, but to the best of my knowledge no sensor/servo activated system. Maybe (probably) I am missing something, but the concept doesn't seem all that difficult to mount and attach a small servo motor to a link arm. Whether it is mounted in a 30 year old TR or new shiny metal seems inconsequential. Thought I would just try to see if any such systems might be out there in the "builder market". Again...THANKS!
>> Edited by 308tr6 on Wednesday 18th January 19:57
>> Edited by 308tr6 on Wednesday 18th January 19:57
Yes, it is entirely possible, not particularly hard to do, and all the parts required are available off-the-shelf. I think you are unlikely to find a complete kit from an automotive supplier though; it'd be a question of looking through industrial automation catalogues for suitable sensors and actuators, and building a small amount of electronics to link them together. The sensors and actuators will not be cheap...
Speaking as one who is of similar mind when it comes to fiddling with stuff on my bike, and being equally comfortable with electronic and mechanical fiddling, the first thing I would look at would be simply using a thicker throttle cable. Perhaps it would be possible to modify a clutch cable, or a brake cable off a bike. Another possibility would be a pushrods-and-bellcranks linkage in the manner of a Volvo Amazon.
Speaking as one who is of similar mind when it comes to fiddling with stuff on my bike, and being equally comfortable with electronic and mechanical fiddling, the first thing I would look at would be simply using a thicker throttle cable. Perhaps it would be possible to modify a clutch cable, or a brake cable off a bike. Another possibility would be a pushrods-and-bellcranks linkage in the manner of a Volvo Amazon.
The cable failure is probably due to a misalignment or kink in the inner at some point in the throttle travel. I think it would be a hundred times easier to sort that out than to get an electronic control system working, it will also work better (no lag) and avoid the scary failure modes associated with an electronic throttle control. Make sure your cable is up to the loads you're placing on it, if the throttle return springs are very heavy consider whether they can be lightened, sort out the cable fittings and routing and I think you should be able to sort the problem.
Lots of sensible advice already.
Summary…
Sort obvious problems with cable and return springs. If that’s not done it then.
Try using clutch cable instead as they are designed for heavier loads and bigger springs!
If you really want to do it by wire then there are radio control servo testing circuits, that can be DIY built, which control servo travel using a pot. These are obviously not designed for safety critical applications so inventor beware and also, consider how this might affect insurance/liability! Servos are a little slow travelling from one extreme to the other IMO which might make driving interesting
And if you want another off the wall idea… Why not hydraulics? Clutches have historically gone from cable operated to hydraulic when exported versions required the longer cable runs associated with relocating the driver on the other side.
All the best and make it safe
Leo
Summary…
Sort obvious problems with cable and return springs. If that’s not done it then.
Try using clutch cable instead as they are designed for heavier loads and bigger springs!
If you really want to do it by wire then there are radio control servo testing circuits, that can be DIY built, which control servo travel using a pot. These are obviously not designed for safety critical applications so inventor beware and also, consider how this might affect insurance/liability! Servos are a little slow travelling from one extreme to the other IMO which might make driving interesting
And if you want another off the wall idea… Why not hydraulics? Clutches have historically gone from cable operated to hydraulic when exported versions required the longer cable runs associated with relocating the driver on the other side.
All the best and make it safe
Leo
Thanks for all the comments..a lot of what has been mentioned I have done or messed with, stronger cable, less spring rebound at the carbs etc.,. It is functional and is working right now, but last thing that happened was the little ball at the end of the cable pulled off. I have a pretty strong amount of mechanical advantage and I keep finding weak points. I actually spot welded the little ball end on the cable this last time and it is working, but I never leave home without a cell phone!! Decided to just maybe abandon the cable system for this drive-by-wire, seems a good approach, and I kinda like the "modern tech" aspect of it. If I decide to stick with the cable approach I was actually thinking running 3 cables to my pedal since each carb pair has a single rocker/cable that I currently combine into one pull about 24 inches past the carbs. This device seems to have a lot of drag. The 3 cables wouldn't be as clean looking, but I could route them fairly tidy and zip tie them together. Anyway, thanks for the great responses!
lesliehedley said:
Why don't you try a similar set up to the way Aston Martin control the carbs on their early straight six engines. They use pivoted rods connected to the pedal and the carbs directly. Of course, the carbs would need to be on the same side of the engine as the throttle pedal!
Yeah, this actually plan B or C maybe, however as you probably know the TR6 carbs are on the opposite side of the gas pedal (for LHD cars) but I think I can still direct route linkage. For those of us who drive on the (other side of the road) the standard US TR6 linkage crosses over through a less than perfect method in the firewall, which I abandoned for my current cable setup. Anyway, long winter where I am at, so plenty time to sort out options. Thanks..
Your pedal must be pretty damn heavy if the linkage is strong enough to break the cable.
Im sure a properly manufactured throttle cable would take some force to removed the crimped ends.
www.speedycables.co.uk can make pretty much any cable needed
Im sure a properly manufactured throttle cable would take some force to removed the crimped ends.
www.speedycables.co.uk can make pretty much any cable needed
I'm working on something generic, that would do that. Connect your servomotor to one end, your sensor to the other. Draw a schematic describing what you want in the middle, and download it to the electronics. Job's a goodun.
The only trouble is that it's not ready yet. Still working on the software!
Edited to add;
Are you sure that a PID controller is something that you really want on the throttle. Im sure it can be done, but I'd bet it takes as much patience to set up as a cable solution.
>> Edited by dilbert on Thursday 19th January 17:41
The only trouble is that it's not ready yet. Still working on the software!
Edited to add;
Are you sure that a PID controller is something that you really want on the throttle. Im sure it can be done, but I'd bet it takes as much patience to set up as a cable solution.
>> Edited by dilbert on Thursday 19th January 17:41
ATG said:
that sounds pretty damn cool, dilbert. So you're saying you could do something like draw your PID feedback wotsit and the steady state relationship you want between throttle sensor and servo sensor and your software programmes a microcontroller to do the business?
That's pretty much the size of it.
I have to say I was thinking more about things like electric windows and seats and stuff tbh.
There's quite a lot of people doing engine management systems, and they really aren't that difficult to do "relatively". I'm not saying that engine management is easy, more that you have a map, and the management unit just gets programmed with your particular weights.
With a generic controller you have to provide a way of describing what you wan, create a netlist, minimize it, and work out how to make that configuration happen. In short the software is heavier, but the hardware is simpler.
For those reasons it seemed to be a better thing to try and do.
Just think..... Rather than jerky windows you could have accellerated windows, with memory presets!!!
Maybe indicators that fade, rather than flash.
I don't know, I'm sure there's loads of things you could do!
You could do all kinds of stuff with that. In addition to the feedback and response stuff you could have sanity checks on the sensor readings, self calibration, big flashing red lights when it thinks it has gone tits up. You could even make the heater control in a Chimaera work. If you don't mid me asking, do you have a particular family of microcontrollers in mind? Are you going for something with a bit of computational beef or keeping it as light as possible? Any estimate at all of prices?
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff