may design a custom electronic boost controller, anyone interested?

[wsc]

hey guys, i have been thinking about designing and building a custom electronic boost controller targeted at people who want a nice solid unit for a good price. the idea is you get input on what features are built into the unit and dont have to pay and arm and a leg to get them. i'd really like to know how many people would be interested in this sort of thing.

so what kind of features do you guys want to see? what do you think is mandatory and what would be considered just a cool option? what features do you think are stupid and pointless? right now i'm thinking it will have a nice bright LCD display, a simple knob to adjust the maximum boost level, the ability to show boost in PSI or bar by pressing a button, and the ability to save the highest boost you hit until you press clear or something like that.... oh and it will come in a carbon fiber box so it looks cool

i'm still looking for good parts suppliers right now, and no development has been done. i also dont have a turbo'd car so i will need to find some victims - i mean testers - to help me out ... anyway, hopefully the price can be kept around $200-$250, how does that sound? its early enough that if you speak up, your voice will be heard!

fire away

[sonorous986]

it might be a little too much, but maybe the ability to save two or three presets? like "street boost", "Track boost", etc., so you can push on button and have it set to 5 psi, then when you're at the track hit another button and have it up to 10 or something.

[bruinbear714]

hey guys, i have been thinking about designing and building a custom electronic boost controller targeted at people who want a nice solid unit for a good price. the idea is you get input on what features are built into the unit and dont have to pay and arm and a leg to get them. i'd really like to know how many people would be interested in this sort of thing.

so what kind of features do you guys want to see? what do you think is mandatory and what would be considered just a cool option? what features do you think are stupid and pointless? right now i'm thinking it will have a nice bright LCD display, a simple knob to adjust the maximum boost level, the ability to show boost in PSI or bar by pressing a button, and the ability to save the highest boost you hit until you press clear or something like that.... oh and it will come in a carbon fiber box so it looks cool

i'm still looking for good parts suppliers right now, and no development has been done. i also dont have a turbo'd car so i will need to find some victims - i mean testers - to help me out ... anyway, hopefully the price can be kept around $200-$250, how does that sound? its early enough that if you speak up, your voice will be heard!

fire away

I have thought about this and it isn't too difficult, considering my electrical engineering background. However, it'd be hard to find people who will be willing to try out the controller without guarantee & proof that it won't misbehave (overboost, spike, etc etc). That, and it'd be expensive to get a custom lcd screen unless you stick with the standard lcds, in which case won't be the same size as 99% of the boost controllers out in the market.

Now on with features - if I were to design the boost controller, I would have the ability to adjust the parameters of the control loop to cater to how I want the boost response to be. Another feature would be a built-in turbo timer since the two almost go hand in hand. I could care less about how it looks since it'll semi-hidden from view.

[wsc]

it might be a little too much, but maybe the ability to save two or three presets? like "street boost", "Track boost", etc., so you can push on button and have it set to 5 psi, then when you're at the track hit another button and have it up to 10 or something.

that should be pretty easy to do actually...

I have thought about this and it isn't too difficult, considering my electrical engineering background. However, it'd be hard to find people who will be willing to try out the controller without guarantee & proof that it won't misbehave (overboost, spike, etc etc). That, and it'd be expensive to get a custom lcd screen unless you stick with the standard lcds, in which case won't be the same size as 99% of the boost controllers out in the market.

Now on with features - if I were to design the boost controller, I would have the ability to adjust the parameters of the control loop to cater to how I want the boost response to be. Another feature would be a built-in turbo timer since the two almost go hand in hand. I could care less about how it looks since it'll semi-hidden from view.

Good suggestions! I had never thought of letting the user adjust the parameters like that. For testing, the unit will be subjected to compressed air from a regular compressor to make sure it responds well before it ever goes onto a vehicle. I hope this testing idea will work, but if anyone has a better idea, I would love to hear it. As far as the LCD, yes I am having trouble locating an LCD of the "right" size, but I will keep scouring until I find something that will somehow work. If that doesnt work out, I guess I could always go back to an LED display =/

Could anyone post the dimensions "most" boost controllers are? Would it be bad if this unit was a different size?

I was thinking about more features: how does everyone feel about datalogging? I wonder how much something like that would be used? How about some kind of user specified "overboost" warning? Like a pulsating LED that goes off to remind you when you are nearing your car's "safe limit" so you dont go overboard cranking the boost up. Or, maybe it could be set up such that when you enter the "programming" mode on the unit, you can change the maximum possible boost level so you NEVER exceed it...

Keep the thoughts coming guys

[bruinbear714]

that should be pretty easy to do actually...

Good suggestions! I had never thought of letting the user adjust the parameters like that. For testing, the unit will be subjected to compressed air from a regular compressor to make sure it responds well before it ever goes onto a vehicle. I hope this testing idea will work, but if anyone has a better idea, I would love to hear it. As far as the LCD, yes I am having trouble locating an LCD of the "right" size, but I will keep scouring until I find something that will somehow work. If that doesnt work out, I guess I could always go back to an LED display =/

Could anyone post the dimensions "most" boost controllers are? Would it be bad if this unit was a different size?

I was thinking about more features: how does everyone feel about datalogging? I wonder how much something like that would be used? How about some kind of user specified "overboost" warning? Like a pulsating LED that goes off to remind you when you are nearing your car's "safe limit" so you dont go overboard cranking the boost up. Or, maybe it could be set up such that when you enter the "programming" mode on the unit, you can change the maximum possible boost level so you NEVER exceed it...

Keep the thoughts coming guys

Hooking it up to an air compressor probably isn't the best way to do it. The actual in-car response of the turbo is going to be different than opening a valve and having air rush in from the air compressor.

Finding a bubblegum stick sized LCD is going to be tough. Good luck with that. I already looked and havn't found any available for purchase in single quantities.

Datalogging - forget about that. No one is going to hook up a laptop to the boost controller just to see how the boost response curve looks like.

Overboost - that's a nice feature. This is the steady state boost setting we're talking about right? When the turbo is spooling, I WANT the system to overboost by a few psi and then settle back to a set value to allow the system to reach that value quicker. However, I don't want the system to creep past the set value by more than X amount of psi.

Pulsing an LED - again, the controller will not be in the vision region, so instead of pulsing an LED, try setting off a small speaker, like the kinds used in smoke detectors.

Good luck with it.. it's not as easy as it seems. Designing a basic boost controller is one thing, but designing one that behaves as expected with user definable parameters is much more difficult.

[wsc]

Thanks for the thoughts and advice bruinbear.

One quesion I have for you is, when you say adjustability of the parameters, would you mean, for example, the coefficients in a PID loop? Do you think a PID loop is appropriate for this sort of control? Besides the "X" PSI of boost creep as you posted above, what other parameters would you say are desirable?

I'm starting to come to terms with how difficult this will be, but I'm still in college so I think this would be a good product/project to work on for alot of reasons. This may be too difficult, or it may not. Or I may decide to stick with a very basic level of functionality (initially) and do free firmware updates for all customers as more features are designed and tested.

As far as the LCD goes, I just realized I have a perfect one sitting right in front of me... too bad its in my motorola t720 phone maybe I can find the part number...

One last thought, does anyone have a preference for control knobs? Potentiometer (smooth, but limited rotation) or encoder (clicks as it turns, turns either direction infinitely)?

[bruinbear714]

Thanks for the thoughts and advice bruinbear.

One quesion I have for you is, when you say adjustability of the parameters, would you mean, for example, the coefficients in a PID loop? Do you think a PID loop is appropriate for this sort of control? Besides the "X" PSI of boost creep as you posted above, what other parameters would you say are desirable?

I'm starting to come to terms with how difficult this will be, but I'm still in college so I think this would be a good product/project to work on for alot of reasons. This may be too difficult, or it may not. Or I may decide to stick with a very basic level of functionality (initially) and do free firmware updates for all customers as more features are designed and tested.

As far as the LCD goes, I just realized I have a perfect one sitting right in front of me... too bad its in my motorola t720 phone maybe I can find the part number...

One last thought, does anyone have a preference for control knobs? Potentiometer (smooth, but limited rotation) or encoder (clicks as it turns, turns either direction infinitely)?

Yes, by adjustment of parameters I mean adjusting the coefficients of the PID controller. Using a PID control controller is a given, since you can't control steady state, overshoot, and error with just a PI or PD controller.

With any PID controller design, you have to specify the coefficients and usually this is non-user adjustable to give a predictable response curve. However, in a situation where you have fuel management support for 20psi but would only like to run it at 7psi, it's okay to spike to 15psi and come down to 7psi if the benefit is that the turbo reaches full boost 500rpm sooner.

Physically this is done by keeping the wastegate shut until boost is at 15psi, then opening/closing it to regulate boost at 7psi, rather then cracking it open while the turbo is spooling until 7psi is achieved.

As an analogy, if you had a temperatured controlled room set to 70 degrees, would you be okay if the A/C made the room 65 and then come back to 70 if it takes 30 mins to do so, rather than taking 1 hour to come down to 70 degrees without dipping below it?