I may have had a beer but... DIY Active suspension how hard?
Discussion
I am not aiming for fully computer controlled active suspension but I reckon getting rid of brake dive and corner rolls can't be that hard nowadays.
I'm not looking for any clever sensors that have to read outside of the car. It's simple enough to tell when a brake is being pressed so the car is going to dive forward. Or, when you're steering, you can use inertia sensors to tell if you're going to roll to one side or the other.
If you get rid of brake dive, acceleration squat as well, and any fundamental budy roll of going fast into a corner, I think you remove a whole shed load of problems with handling. It's not like we're talking big distances either. I mean, the move is, I don't know, 5-10 cm. Fit a section to the top of the suspension that could move 5cm-10cm quickly under computer control and I think we've got a starter there.
If you press the brakes:
- you know how hard you press the brakes
- you know how fast you are going
Therefore you know how much the car wants to dive. So raise the front of the car that much and there is no dive,
If you turn the steering wheel:
- you know how fast you're going
- you know how much you turn the steering wheel.
Therefore, you know how much your car is going to roll. Raise the corner of the car, and there is no roll.
So if you have four of these little raisey-upy-downy 5-10 cm pads on each corner, and then a computer that knows how fast you're going, how much you're braking, and how much you're steering, seems pretty straightforward.
I know this doesn't solve every bit of the suspension problem, but it solves an awful lot, very easily and it seems quite simple. Please, what have I got wrong?
Before I go any further, what fundamental things I missed?
Do these already exist for post-fit onto normal cars?
Once these have been calibrated, it seems a really easy post-market fit. Put one of these on top of each suspension strap, and job's a good'un!
I'm not looking for any clever sensors that have to read outside of the car. It's simple enough to tell when a brake is being pressed so the car is going to dive forward. Or, when you're steering, you can use inertia sensors to tell if you're going to roll to one side or the other.
If you get rid of brake dive, acceleration squat as well, and any fundamental budy roll of going fast into a corner, I think you remove a whole shed load of problems with handling. It's not like we're talking big distances either. I mean, the move is, I don't know, 5-10 cm. Fit a section to the top of the suspension that could move 5cm-10cm quickly under computer control and I think we've got a starter there.
If you press the brakes:
- you know how hard you press the brakes
- you know how fast you are going
Therefore you know how much the car wants to dive. So raise the front of the car that much and there is no dive,
If you turn the steering wheel:
- you know how fast you're going
- you know how much you turn the steering wheel.
Therefore, you know how much your car is going to roll. Raise the corner of the car, and there is no roll.
So if you have four of these little raisey-upy-downy 5-10 cm pads on each corner, and then a computer that knows how fast you're going, how much you're braking, and how much you're steering, seems pretty straightforward.
I know this doesn't solve every bit of the suspension problem, but it solves an awful lot, very easily and it seems quite simple. Please, what have I got wrong?
Before I go any further, what fundamental things I missed?
Do these already exist for post-fit onto normal cars?
Once these have been calibrated, it seems a really easy post-market fit. Put one of these on top of each suspension strap, and job's a good'un!
Anti-dive geometry and ARBs are pretty effective without adding active systems. The next step after that is adaptive dampers and then active ARBs. Some of the latest passive frequency-selective dampers are pretty effective too. I don’t see much to gain from DIY active suspension. It will most likely just make it lose grip on anything less than billiard smooth road.
The electrio-mag dampers are solving a different stage of the problem. Use the fancy dampers as well but this idea doesn't affect them.
Remove the need for the anti-dive and roll bars and you free up new options. Your car is now mostly flat, so the ride feels better even though it doesn't affect the loads of the original suspension.
This means you got all the benefits of the original suspension, fancy dampers, everything you want, but your car stays level for the ride as well. The tech involved in doing this is relatively cheap, four pads and a clever computer, so as an aftermarket fit it does seem a viable option.
Remove the need for the anti-dive and roll bars and you free up new options. Your car is now mostly flat, so the ride feels better even though it doesn't affect the loads of the original suspension.
This means you got all the benefits of the original suspension, fancy dampers, everything you want, but your car stays level for the ride as well. The tech involved in doing this is relatively cheap, four pads and a clever computer, so as an aftermarket fit it does seem a viable option.
J4cko - my thoughts exactly. It's really simple to do, technically. It just needs some kind of pattern over a simple processor. I imagine as an aftermarket mod, it would be great. £300, and your car stays stable? Possibly a bit more, depending on how much the kit costs to make, but just seems something we should do.
J4cko - my thoughts exactly. It's really simple to do, technically. It just needs some kind of pattern over a simple processor. I imagine as an aftermarket mod, it would be great. £300, and your car stays stable? Possibly a bit more, depending on how much the kit costs to make, but just seems something we should do.
wca said:
The electrio-mag dampers are solving a different stage of the problem. Use the fancy dampers as well but this idea doesn't affect them.
Remove the need for the anti-dive and roll bars and you free up new options. Your car is now mostly flat, so the ride feels better even though it doesn't affect the loads of the original suspension.
This means you got all the benefits of the original suspension, fancy dampers, everything you want, but your car stays level for the ride as well. The tech involved in doing this is relatively cheap, four pads and a clever computer, so as an aftermarket fit it does seem a viable option.
Are you looking to eliminate all body pitch and roll or just reduce it a bit? If you are going to remove the ARBs then it has a fair bit of lifting to do. Not sure how you are going to change the anti-dive geometry either with your after-market kit. Remove the need for the anti-dive and roll bars and you free up new options. Your car is now mostly flat, so the ride feels better even though it doesn't affect the loads of the original suspension.
This means you got all the benefits of the original suspension, fancy dampers, everything you want, but your car stays level for the ride as well. The tech involved in doing this is relatively cheap, four pads and a clever computer, so as an aftermarket fit it does seem a viable option.
It's not all that hard. It could be done with full pneumatic suspension, a level sensor per wheel, a few accelerometers, some valvegear and an appropriately sized pump. The whole lot could be controlled by a Raspberry Pi or Arduino, in its spare time from playing ECU. Or you could hair-shirt it and use op-amps. Tying in adaptive damping would be a very useful refinement.
Or you could start with the active anti-roll BMW use on some of their 7-series'. Mounting the front bar in a rotatable housing would give you anti-dive; BMW's multi link rear setup does a pretty good job of dealing with squat.
The biggest obstacle is that no-one wants it enough. Cars that are big enough to get a decent ride designed into them don't tend to be driven hard enough that the handling compromises show up*. Cars that people expect sharp handling from are excused the poor ride as part of the experience.
* Yes, I have tracked an E38 750iL, but that was when the track car broke down and the tow car was just sitting there...
Or you could start with the active anti-roll BMW use on some of their 7-series'. Mounting the front bar in a rotatable housing would give you anti-dive; BMW's multi link rear setup does a pretty good job of dealing with squat.
The biggest obstacle is that no-one wants it enough. Cars that are big enough to get a decent ride designed into them don't tend to be driven hard enough that the handling compromises show up*. Cars that people expect sharp handling from are excused the poor ride as part of the experience.
* Yes, I have tracked an E38 750iL, but that was when the track car broke down and the tow car was just sitting there...
You can't do it with set parameters - this is where it starts getting complicated (ish)
You say you put your foot on the brake and you know how much the car will tilt forward - you don't, the loading (people and stuff) as well as the positioning of the load will change this journey by journey - also the amount of grip, tyre pressures, the age / condition of the suspension - how much fuel your carrying will all make a difference - therefore you have to have external sensors.
Imaging braking on ice - car sees you brake and jacks the front up, but the car doesn't transfer as much weight, so it goes nose high.
The second part is show me actuators that can respond fast enough, that will deal with salt / grime etc over years and years reliably - for a sensible amount of money - anything that will respond quickly enough will be operating at very high pressures - with all the complexity of pumps and hoses etc.
I think the systems such as those found on mclarens are as good as we'll get for now.
You say you put your foot on the brake and you know how much the car will tilt forward - you don't, the loading (people and stuff) as well as the positioning of the load will change this journey by journey - also the amount of grip, tyre pressures, the age / condition of the suspension - how much fuel your carrying will all make a difference - therefore you have to have external sensors.
Imaging braking on ice - car sees you brake and jacks the front up, but the car doesn't transfer as much weight, so it goes nose high.
The second part is show me actuators that can respond fast enough, that will deal with salt / grime etc over years and years reliably - for a sensible amount of money - anything that will respond quickly enough will be operating at very high pressures - with all the complexity of pumps and hoses etc.
I think the systems such as those found on mclarens are as good as we'll get for now.
J4CKO said:
I want to know what happened to that Bose system that was tested on Lexus LS400s back in the day, looked utterly amazing, but was canned for being too expensive and required too much processing, can probably be done with a Raspberry Pi nowadays.
It’s on sale. https://www.clearmotion.com/
esuuv said:
You can't do it with set parameters - this is where it starts getting complicated (ish)
You say you put your foot on the brake and you know how much the car will tilt forward - you don't, the loading (people and stuff) as well as the positioning of the load will change this journey by journey - also the amount of grip, tyre pressures, the age / condition of the suspension - how much fuel your carrying will all make a difference - therefore you have to have external sensors.
Imaging braking on ice - car sees you brake and jacks the front up, but the car doesn't transfer as much weight, so it goes nose high.
The second part is show me actuators that can respond fast enough, that will deal with salt / grime etc over years and years reliably - for a sensible amount of money - anything that will respond quickly enough will be operating at very high pressures - with all the complexity of pumps and hoses etc.
I think the systems such as those found on mclarens are as good as we'll get for now.
ZF Smotion, on sale on some Audis and Porsches. I don’t think McLaren offer an active suspension, but I could be wrong.You say you put your foot on the brake and you know how much the car will tilt forward - you don't, the loading (people and stuff) as well as the positioning of the load will change this journey by journey - also the amount of grip, tyre pressures, the age / condition of the suspension - how much fuel your carrying will all make a difference - therefore you have to have external sensors.
Imaging braking on ice - car sees you brake and jacks the front up, but the car doesn't transfer as much weight, so it goes nose high.
The second part is show me actuators that can respond fast enough, that will deal with salt / grime etc over years and years reliably - for a sensible amount of money - anything that will respond quickly enough will be operating at very high pressures - with all the complexity of pumps and hoses etc.
I think the systems such as those found on mclarens are as good as we'll get for now.
The electronic bits are easy. Consider stability management systems that will sense the start of a spin and catch the car for you before you even realise there is a problem. Sensors and ecus are reasonably cheap, super compact, reliable, react in milliseconds.
It's the mechanical bits that are difficult. Anything that just stiffens a damper (easy) rather than increase spring preload (difficult) to prevent dive, squat, or roll, by definition reduces suspension movement making handling sketchy and skittish and can induce under or oversteer.
Difficult is not impossible. Just costly, and adds weight and complication.
It's the mechanical bits that are difficult. Anything that just stiffens a damper (easy) rather than increase spring preload (difficult) to prevent dive, squat, or roll, by definition reduces suspension movement making handling sketchy and skittish and can induce under or oversteer.
Difficult is not impossible. Just costly, and adds weight and complication.
996TT02 said:
The electronic bits are easy. Consider stability management systems that will sense the start of a spin and catch the car for you before you even realise there is a problem. Sensors and ecus are reasonably cheap, super compact, reliable, react in milliseconds.
It's the mechanical bits that are difficult. Anything that just stiffens a damper (easy) rather than increase spring preload (difficult) to prevent dive, squat, or roll, by definition reduces suspension movement making handling sketchy and skittish and can induce under or oversteer.
Difficult is not impossible. Just costly, and adds weight and complication.
Isn't there an aftermarket system for retro fitting to Porsches called TractiveIt's the mechanical bits that are difficult. Anything that just stiffens a damper (easy) rather than increase spring preload (difficult) to prevent dive, squat, or roll, by definition reduces suspension movement making handling sketchy and skittish and can induce under or oversteer.
Difficult is not impossible. Just costly, and adds weight and complication.
https://ttsuspension.co.uk/product-category/brand/...
ATM said:
Kawasicki said:
ZF Smotion, on sale on some Audis and Porsches. I don t think McLaren offer an active suspension, but I could be wrong.
I thought that's McLaren's thing their active suspension instead of anti roll bars.Gassing Station | Suspension, Brakes & Tyres | Top of Page | What's New | My Stuff