65 replies to this topic

[David128]

I had a thought re idea. new drivers should be taught in a car without any electronic aids. No ABS No Power steering. No braking round corners. And maybe a skid pan. Just my opinion, but how many times have you panicked. over corrected and made things worse. Experience counts and you only get it by doing it.

[Dylan8660]

Spiguy, I think you might be elevating me up too much, I’m probably more of a dancing guy on this one.

 

 

There is very little to disagree with on any of the posts made, they all have some validity to what I have proposed. I am fully aware that the proposal is a crude blunt instrument and a lot of the discussion has revolved around refining something that doesn’t exist but it’s important to have that discussion outlining the difficulty of having a more preferable refined system as it has clarified why the proposal is so crude even though it has probably clouded its intended purpose. I am really just trying to work with what is already in place.

 

As far as describing in technical terms, I don’t think the operating principles have been discussed enough to move onto that yet. What I mean by operating principles is examining what is currently in place, how it is used, the circumstance when that fails, the uniqueness of that circumstance and if that can be used as the system trigger, would replicating what should happen work better than the current system, could it ever be dangerous and if any of that, when understood, be even desirable.

 

I’ve no timescale to work to, nor am I craving agreement. It’s been fascinating to see what people have picked out and interpreted from what’s been said. I don’t know if I can add any more to this but I’ll have a think over the weekend. I’m enjoying the current discussion on driving skills though, again very valid.

[monkey]

 
As far as describing in technical terms, I dont think the operating principles have been discussed enough to move onto that yet. What I mean by operating principles is examining what is currently in place, how it is used, the circumstance when that fails, the uniqueness of that circumstance and if that can be used as the system trigger, would replicating what should happen work better than the current system, could it ever be dangerous and if any of that, when understood, be even desirable.

This has been discussed - there is no reliable way, using what is current in place to decide when additional help is required. The only things you could measure, using what is in place is brake line pressure or brake peddle position. And as has been said, there are so many external factors (road surface/weather/tire wear/suspension setup) which will change the brake position or brake line pressure required to lock the wheels.

Therefore as has already been said, the only way to detect when wheels lock up would be motion sensors on the wheels.

[Ethel]

...the best way certainly. You could measure measure deceleration, or torque reaction, against brake pressure. Even engine revs could give some sort of signal, but nothing will be as simple and reliable as wheel motion. Though this did start as cadence and specifically not ABS, so we could ask if pulsing in extra pressure once you've physically pushed the pedal as far as you can could be of any benefit?

 

It's the activation that presents the real challenge if you don't want to risk having less stopping power than before.

[nicklouse]

But what does an ABS system do?

 

it replicates cadence braking.

 

the wheel is already round.

[tiger99]

I strongly advise you not to proceed with this project. Some time ago I was professionally involved with ABS systems, actually on aircraft, which are much easier because they have a powered hydraulic system. I can tell you that the design and development effort is immense, then comes the safety aspects like verifying and validating the software. It can not, repeat, can not, be done at home. You need a large team (15 to 20) of the right people, with the right facilities.

 

A Raspberry Pi will not do the job. You need a safety critical processor of some sort, with all sorts of independent backups. I am not even going to begin to describe how to do it because it will only encourage someone without the requisite knowledge to have a go, and probably kill someone.

 

Frankly, the only hope is to adapt a complete system from another small car, and even that is extremely difficult because you will not have access to the source code of the software, and it is completely idiotic to go blindly hacking into binary code in a safety critical system.

 

I can guarantee that a home-made ABS system will invalidate your insurance, and there is about a 99% chance of having a spectacular accident.

 

Sorry, but there are some things that you can not, and must not, do at home. This is one.

 

But, in a Mini you would mostly need ABS on the rear wheels, as the fixed pressure limiting valve cannot be correct for every combination of load and road surface. A load sensing pressure limiting valve off another vehicle would help, but you would need at least a thin anti-roll bar, preferably not of the amateurish design from you know who, because the takeoff point for the valve is the middle of the ARB, where it sees the average vertical position of both wheels. This is a not very good passive system, no software etc, but may be better than the fixed limiter. But your insurance company might not like that one either. Personally I think it is worth a try, as loss of rear brakes (only do it in a dual circuit system!) is often not as serious in a Mini as locked rear wheels.