Sim Brake Thoughts

[pilmat]

I wonder if it's somehow possible to use a real ABS system somehow with the game output.
That is if one would get really fancy with their brakes.

You'd have to have 1 master cylinder, ABS pump, valve system, controller and 4 slave cylinders.

You'd then need to use something like an Arduino or something to send game data to the controller of the pump to 'fool' it into thinking a real car is at work.
So you'd have to send the data of each wheel's speed to the controller. At which point it will work out for itself if it needs to activate.

I believe Fanaleds already uses ABS output of various games. Though I don't know if this is on a per tire basis.

This does mean one thing though. If it would work, you could switch off your 'emulated ABS' in the game's options/aids menu and use a switch for example to turn on your 'real' ABS.
Thus in return getting very different results and moving the functions from the game code to the input device. Like Leo Bodnar talked about (though with FFB systems) in his FFB wheel article.

Sorry, just talking wild for a moment here to maybe give someone (or maybe you Pilmat) a new idea.

I've copied this over from the Hydraulic Brake Build thread, as I think it is quite relevant and should stay in this stream of thought

ABS is a very complex system designed to tackle a very complex problem. Simply put, ABS turns a single input (pedal pressure) into an independent 4 channel output. The system breakdown you listed is missing one of the key elements of the system's function: the 4 separate wheel speed sensors to determine when one of the wheels is slowing down faster than the others (and therefore requiring a different mount of braking). Here the game would require 4 wheel speed outputs to the external ABS controller, etc.

My first question reading this was "what does he want to achieve that is not already covered in the game?" Our external sim toys have to give us a tactile control over something we normally control, but as ABS is really a car parameter (and the most sophisticated do simulations on board to handle the real world input) why would you want this controlled external to the "car" (i.e. simulator)? If I can see what you want to achieve, we can discuss it's benefits and drawbacks

 

[Mayaman]

Any links for Century Spring?

http://www.centuryspring.com/?utm_s...aign=brBrand&gclid=CLXm1rum2bUCFQP0nAodszYA5w

 

[LogiForce]

I believe I did mention tyre speed needing to be send to the controller.

As to why?
The feel is real and not some toe tickling vibration motor you wouldn't feel in the het of battle (at least I don't).
Making ABS an external input so to say will give the end user more control over the ABS system, if the controller is a designed one for sim racing purposes.
The external ABS system can easily become far more sophisticated than that of any game's internal ABS system, which could give the driver another edge and another thing to tweak.

Drawbacks? Sure.
As you mentioned not every sim will output speed per tyre, they might only send a lockup signal per tyre. Or even just a general signal.
The ABS behavior via original controller might not cause the desired effect because the simulated car breaks either react too quickly, have a different throw or linearity and so on.
The initial work having to go into the controller and possibly reverse engineer a real ABS system will take a lot of time and effort to get an initial working product with new specialized contoller.

 

[ravey1981]

How many race cars have ABS though? Isn't it in with the junk thats chucked out when turning a road going car into a race car?

 

[LogiForce]

How many race cars have ABS though? Isn't it in with the junk thats chucked out when turning a road going car into a race car?

That's one of the reasons why I never use ABS, as I rather hone my braking skills while staying pure to the sim.

That said. How many DO in fact use the clubsport pedals ABS motor as they find it hard to find the threshold?
Now depending on the answer it can be used as a 'cheat' or as a reason to use ABS more properly in their sim.

 

[pilmat]

That's one of the reasons why I never use ABS, as I rather hone my braking skills while staying pure to the sim.

That said. How many DO in fact use the clubsport pedals ABS motor as they find it hard to find the threshold?
Now depending on the answer it can be used as a 'cheat' or as a reason to use ABS more properly in their sim.

ABS is like traction control, most low end stuff (i.e. mass produced road cars) is almost a hindrance. But high end race stuff is awesome, there wouldn't be rules AGAINST it if it didn't work

Ripping it out of a road car for performance makes a lot of sense, especially when you realize how many more meters of brake line are added by it. Like removing the booster and putting on a proper 2 channel brake system.

But high end systems do what our sim should be doing, analyze what the car is doing, dependent on track position, and act accordingly. I remember watching a DTM spin in 1995 with the ABS system, it worked each individual channel to allow the driver to recover from the spin without stopping. It was AMAZING to watch!!!

I go back to my initial question, what does a sim not do that we think could be improved with an external system? If it is just to add tactile control to the brake pedal for impending lock-up, then we could tackle that problem completely differently than with an external ABS system. A simple shaker with the appropriate output from the sim would do just fine. This is now well out of my league (I'm mechanical ), folks like DSD and Mr.Basher can chime in and tell you if I'm nuts

I can now see what LogiForce is after, but I think there is a much simpler way to get there

 

[LogiForce]

Eh, shakers/vibration motors do not feel like the pressure difference caused by an ABS system though.
To be honest, if it's pure vibration it wouldn't even be worth the money or effort. Especially as I rather reproduce this via the Simvibe chassis system, as you can at that point just feel the tyres losing traction (I reckon, as I just have a sile BKG2).

 

[left888]

ABS is like traction control, most low end stuff (i.e. mass produced road cars) is almost a hindrance. But high end race stuff is awesome, there wouldn't be rules AGAINST it if it didn't work

This is true to a degree, you look at the new McLaren MP4-12C GT3 we got at iRacing. It certainly is high end and has ABS brakes, however that car is considered more of a gentleman's race car. Something like getting a dozen rich dudes driving around the race track.

Where the power cars say like the DP have no ABS and they are fast and brake points are so much different then say the McLaren.

 

[kikie]

Reviving this thread because I bought the T300RS and would like to upgrade the brake pedal to a more progressive brake pedal which can be modulated perfectly. This is the only thing I'm after.

I re-read this thread and I'm still nog convinced that an Load Cell or hydraulic brake system is worth the money, except when it is possible to make a DIY hydraulic brake without a pressure sensor.

What I want to do is get a master brake cylinder and a brake caliper or a drum brake. If I should use a brake caliper, I'm not going to use a disc brake. I want to use a softer material, such a rubber block to insert in the brake caliper, reduce the brake stiffness and increase the pedal travel.

Only problem that I have that I still want to use the pot instead of a pressure sensor. Which means that I'm probable not going to use the full deflection (don't know if this is the correct word) of the pot.

If I calibrate the brake pedal with a pot and reduced travel in windows and in-game, do I still have the same result as when I use the full distance to the pot when using a stock brake pedal with a spring?

It doesn't have to be an hydraulic brake or a load cell, as long as the feel of the pedal is progressive and can be modulated perfectly.

Muscle memory what I'm after. With a standard brake pedal (spring), you have to remember how far you have to push the brake pedal to not lock the wheels. Muscle memory is easier and more natural to modulate the brake pedal.

 

[Jet Badger]

Reviving this thread because I bought the T300RS and would like to upgrade the brake pedal to a more progressive brake pedal which can be modulated perfectly. This is the only thing I'm after.

I re-read this thread and I'm still nog convinced that an Load Cell or hydraulic brake system is worth the money, except when it is possible to make a DIY hydraulic brake without a pressure sensor.

What I want to do is get a master brake cylinder and a brake caliper or a drum brake. If I should use a brake caliper, I'm not going to use a disc brake. I want to use a softer material, such a rubber block to insert in the brake caliper, reduce the brake stiffness and increase the pedal travel.

Only problem that I have that I still want to use the pot instead of a pressure sensor. Which means that I'm probable not going to use the full deflection (don't know if this is the correct word) of the pot.

If I calibrate the brake pedal with a pot and reduced travel in windows and in-game, do I still have the same result as when I use the full distance to the pot when using a stock brake pedal with a spring?

It doesn't have to be an hydraulic brake or a load cell, as long as the feel of the pedal is progressive and can be modulated perfectly.

Muscle memory what I'm after. With a standard brake pedal (spring), you have to remember how far you have to push the brake pedal to not lock the wheels. Muscle memory is easier and more natural to modulate the brake pedal.

If you're going as far as a hydraulic system, might aswell throw a loadcell or a pressure sensor in there. The problem with using pots and some sort of spring material is getting the right force vs deflection relation so that the distance the pedal travels is constantly proportional to the force it is pressed with. A hydraulic system on it's own will do nothing for that, you'll still need to stick some rubber stuff for the slave cylinder to press onto and if you're going with a pot, it's pretty much hit or miss whether you can get it right.

Also using only a part of pot's range will have you lose resolution. It will most likely calibrate and work just fine but instead of possible 255 (or whatever the resolution on that set is) registrable pedal positions you'll have only, let's say half that if you use half of the pot's range.

I'd say if you want to go with a hydraulic setup, you're much better off getting a simple bathroom scale load cell, sticking in a brake caliper, you can solder up the INA122 amp for it no problem and do it like that.

 

[kikie]

Losing resolution of the pot is what worries me a little bit. I was aware of this problem.

Sticking a load cell in the brake caliper is what have been thinking about as well but it is too difficult for me because I don't know anything about electronics. I need a pcb (electronics) to make the load cell work so I can use the pedal playing games. That is the main reason I wanted to keep the pot.

 

[Jet Badger]

Losing resolution of the pot is what worries me a little bit. I was aware of this problem.

Sticking a load cell in the brake caliper is what have been thinking about as well but it is too difficult for me because I don't know anything about electronics. I need a pcb (electronics) to make the load cell work so I can use the pedal playing games. That is the main reason I wanted to keep the pot.

You don't necessarily have to make a PCB for it. A piece of perfboard or veroboard would do just fine for something as simple as this and you can get those in any electronics parts store.

 

[kikie]

I mean a controller board (12-bit?). Or a load cell amplifier.

 

[Jet Badger]

I mean a controller board (12-bit?). Or a load cell amplifier.

But if you wanted to use a controller board, you'd need one not depending on whether you use a pot or a loadcell. And the loadcell amplifier consists just of a handful of components, certainly doable on a perfboard.

 

[kikie]

But if you wanted to use a controller board, you'd need one not depending on whether you use a pot or a loadcell. And the loadcell amplifier consists just of a handful of components, certainly doable on a perfboard.

As I said I don't know anything about electronics. It is impossible for me to make a controller myself using this perfoard thing.

Yesterday I saw the Derek thread and it seems that he is selling a controller board. If I do this, I need to find found out the correct load cell I could use with the setup I mentioned yesterday.

Master brake cylinder (small bore) -----> brake caliper -----> some sort of material that can be compressed (2 parts) inside the brake caliper -----> Load cell in between.

 

[Jet Badger]

Yes, making a controller board would be much harder. I was talking about the load cell amplifier in the past posts - two separate circuits. A separate loadcell amp is just three components including the chip. But that's irrelevant if you're planning to get a retail controller with an amp already included.

Any loadcell woulld do in principle. The only difference is pretty much that there are 4-wire and 3-wire loadcells. 4-wire ones measure load in two directions (the industrial loadcells you see in most DIY projects). 3-wire loadcells measure load in only one direction (Clubsport pedals use those, can be found in bathroom scales). For this purpose a 3 wire loadcell is enough as you want to measure the load in only one direction anyway but you'll just need a couple of resistors to simulate the other "half" that a 4-wire loadcell would have as the 4th wire.

Also there's this problem of how you're gonna put the loadcell in the caliper. The industrial loadcells most use look like this:

In order for them to work one end needs to be anchored and you push/pull on the other end of it. Applying it to work with a brake caliper would be rather tricky.

A regular 3-wire CSP style loadcell:

For this one to work, you pretty much press on it like a pancake. Sticking it into the caliper with some spacing material would do the job. They're also much cheaper.

If you'll be going with junkyard parts, I'd suggest looking into cluch master cylinders rather than brake. They'll have less stuff to go around.

Also check this thread out if you haven't. It's pretty similar to what you have planned. http://www.varjanta.com/forum/hydraulic-brake-pedal-handmade-t1808.html

 

[kikie]

Thanks for the DIY link.

I already saw that varjanta thread about three years ago.

 

[JohnBodin]

IMO, having a hydraulic setup that moves a caliper to depress a load cell is a horribly Rube Goldberg-esque design when it comes to building a pressure-sensitive brake for sim racing (e.g., ridiculously complex and over-designed for what it is intended to do).

If you want hydraulics, then just add a pressure transducer to the hydraulic line and have that read the hydraulic pressure directly. I think hydraulics are vastly overrated for home sim racing use because inevitably any pressurized system filled with oil or fluid is going to leak, and finding a puddle of fluid under your sim racing rig is NOT a nice surprise, especially if carpeting in a habitable area of your household is involved (that is, if you race in your den, family room, office, loft area, etc.).

Using a simple beam-type load cell (similar to the one used in CST pedals) provides you with a direct means to sense the pressure applied to the brake pedal. These type of load cells are generally maintenance-free, and if you choose the proper specs, they may outlast your sim racing hobby (Fanatec's little "bathroom scale" pressure sensors are VERY "weak" components -- that's why they fail, but thankfully they're cheap so they're at least easy ind inexpensive to replace).

The main thing you need with a beam-type load cell is some means of coupling the pedal to the load cell in a manner that allows for some pedal travel in order to make the braking inputs more "convincing." You CAN implement a beam-type load cell that will leave you with no pedal travel, and that kind of install can be VERY precise, but it usually won't "feel" right. Which is why most people use springs or rubber bushings and such to transfer pedal pressure to the load cell.

Just my $0.02.

 

[JohnBodin]

FWIW, if you go the hydraulics + pressure transducer route, this seems to be the transducer most people have used for their home-built hydraulic pedals:

Link: http://www.ebay.com/itm/Pressure-tr...l-air-water-/271511515207?hash=item3f3758f047

 

[kikie]

IMO, having a hydraulic setup that moves a caliper to depress a load cell is a horribly Rube Goldberg-esque design when it comes to building a pressure-sensitive brake for sim racing (e.g., ridiculously complex and over-designed for what it is intended to do).

If you want hydraulics, then just add a pressure transducer to the hydraulic line and have that read the hydraulic pressure directly. I think hydraulics are vastly overrated for home sim racing use because inevitably any pressurized system filled with oil or fluid is going to leak, and finding a puddle of fluid under your sim racing rig is NOT a nice surprise, especially if carpeting in a habitable area of your household is involved (that is, if you race in your den, family room, office, loft area, etc.).

Using a simple beam-type load cell (similar to the one used in CST pedals) provides you with a direct means to sense the pressure applied to the brake pedal. These type of load cells are generally maintenance-free, and if you choose the proper specs, they may outlast your sim racing hobby (Fanatec's little "bathroom scale" pressure sensors are VERY "weak" components -- that's why they fail, but thankfully they're cheap so they're at least easy ind inexpensive to replace).

The main thing you need with a beam-type load cell is some means of coupling the pedal to the load cell in a manner that allows for some pedal travel in order to make the braking inputs more "convincing." You CAN implement a beam-type load cell that will leave you with no pedal travel, and that kind of install can be VERY precise, but it usually won't "feel" right. Which is why most people use springs or rubber bushings and such to transfer pedal pressure to the load cell.

Just my $0.02.

Thanks, your less-than-two cents are worth a lot.

I'm not going to build or buy an hydraulic system anymore after reading your post. I'm not even convinced that a load cell is the correct way either, for me personal.

The only thing I'm interested in, as I said in my first post when reviving this thread, is the correct feel of the brake pedal. I want a brake pedal that can be modulated. It doesn't have to be a 100% realistic feeling as in a real car. All I want is to be able to use the brakes of a car in-game without locking the brakes everytime. And I don't want to play race games were I have to use the ABS on every car.

If a progressive spring or a rubber conical brake mod (T3PA pedals has this mod) does the trick, than I'm all for it. Only problem is the pot which will lose resolution.

At the moment I'm still using my SRW S1 steering wheel. I attached a piece of some sort of rubber foam to the brake peddle (not pedal in this case ). It helps a lot. I'm not locking the brakes that often anymore. So a simple solution is good enough for me as long as I enjoy my race games.

 

[Dean Ogurek]

The use of Pneumatic cylinders has also proven quite effective to replicate realistic feeling pedal resistance/travel; they're also much cheaper than hydraulic components and somewhat simpler in design and installation. If I switch from load-cell based brake (CST currently), I'll go the Pneumatic route. Despite common concerns with leak-down issues, they are proving to be quite reliable; the weakest link is likely to be the pressure-sensor.

I think a hand-brake using a Pneumatic system might be interesting too.

 

[Mr Latte]

Never mind...