Cheap Electronic FCD - Revised
Moderators: Helpinators, Moderators
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
Cheap Electronic FCD - Revised
A little while ago boostjunkie asked me to make a simple FCD for him that would just clamp the pressure sensor reading. I thought I had a good design for a while but it just wasn't working, but I figured it out now. It's pretty simple and very inexpensive.
Here's the design:
http://www.surrealmirage.com/vrg3/fcd/
Here's the design:
http://www.surrealmirage.com/vrg3/fcd/
Last edited by vrg3 on Tue Jun 09, 2009 8:28 pm, edited 2 times in total.
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
Pretty cool stuff.....brings me back to my college days of messing with op amps and stuff
I'm assuming this circuit/design would tell the ECU that pressure is always slightly under its cutoff point correct i.e. always on boost? I think a FCD or something that kicks in only when input voltage from the sensor reaches a certain point would be better. Then again, I'm not super familar if the ECU uses this input for other variables, etc.
What sort of tolerances do these componants have? Could things vary enough in the componants so the FCD doesn't work properly. Also, resistances and such can change slightly if the componants heat up right?
I'm assuming this circuit/design would tell the ECU that pressure is always slightly under its cutoff point correct i.e. always on boost? I think a FCD or something that kicks in only when input voltage from the sensor reaches a certain point would be better. Then again, I'm not super familar if the ECU uses this input for other variables, etc.
What sort of tolerances do these componants have? Could things vary enough in the componants so the FCD doesn't work properly. Also, resistances and such can change slightly if the componants heat up right?
Josh
surrealmirage.com/subaru
1990 Legacy (AWD, 6MT, & EJ22T Swap)
2020 Outback Limted XT
If you need to get a hold of me please email me rather then pm
surrealmirage.com/subaru
1990 Legacy (AWD, 6MT, & EJ22T Swap)
2020 Outback Limted XT
If you need to get a hold of me please email me rather then pm
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
Op amps are the shizzle.Legacy777 wrote:Pretty cool stuff.....brings me back to my college days of messing with op amps and stuff :)
Nooo... If you wanted to do that you'd just need the 10K and 47K resistors and nothing else, not even the pressure sensor. But the ECU would never let you get away with that. Even if it didn't see a problem with manifold pressure being 12.5 psig at idle, it would see a problem with atmospheric pressure being 27 psia. Remember that it uses the same sensor to measure atmospheric pressure.I'm assuming this circuit/design would tell the ECU that pressure is always slightly under its cutoff point correct i.e. always on boost?
Yup, that's what this guy does. Until around 12.5 psi, the ECU sees an accurate pressure signal. Above 12.5 psi, it continues to read 12.5 psi. In fact, if you used this FCD on a completely stock correctly-functioning Legacy Turbo, the ECU would see the exact same signals as it always had.I think a FCD or something that kicks in only when input voltage from the sensor reaches a certain point would be better.
Modulo, of course, the offset voltages in the amplifiers.
Like I said, I was going for cheap and simple, so I calculated it out so as long as you use +/- 10% resistors or better (10% are considered the cheap kind anyway) it would work. You just might end up clamping at a slightly lower pressure depending on how it works out. It shouldn't ever come out higher than 13psi, unless I did something wrong. :)What sort of tolerances do these componants have? Could things vary enough in the componants so the FCD doesn't work properly.
Yes, that is true, depending on the types of resistors. If you really wanted it to be very precise you could use high-power high-precision metal film resistors, but it's not that big a deal. It would only matter in the voltage divider, and in that case what really matters is the ratio of resistances, which won't change as much as the resistance values themselves.Also, resistances and such can change slightly if the componants heat up right?
Ideally an FCD wouldn't eliminate fuel cut but rather raise it to a higher boost level. This one doesn't do that but is very simple and cheap. boostjunkie in particular is exceeding the range of the stock pressure sensor so there isn't much choice in terms of a simple electronic fuel cut.
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
cool,
Thanks for all the clarifications, and yes this little guy looks like it would work quite well and is pretty inexpensive.
Thanks for all the clarifications, and yes this little guy looks like it would work quite well and is pretty inexpensive.
Josh
surrealmirage.com/subaru
1990 Legacy (AWD, 6MT, & EJ22T Swap)
2020 Outback Limted XT
If you need to get a hold of me please email me rather then pm
surrealmirage.com/subaru
1990 Legacy (AWD, 6MT, & EJ22T Swap)
2020 Outback Limted XT
If you need to get a hold of me please email me rather then pm
vr, can you make a simple circuit that will bypass this if the voltage goes beyond a certain threshold, say whatever it is at 16 p.s.i.? (I know boostjunkie's would have to be higher than that...)
that way the fuel cut will hit at a pre-determined but higher level. i don't know how hard something like this would be to make.
that way the fuel cut will hit at a pre-determined but higher level. i don't know how hard something like this would be to make.
1993 WMP BC6 5MT EJ22T 9psi 3.9:1 213k 205/55R16
62.6 m/s @ 0.66 bar. Gotta love boost. :)
62.6 m/s @ 0.66 bar. Gotta love boost. :)
-
- Knowledgeable
- Posts: 5203
- Joined: Thu Nov 07, 2002 3:59 am
- Location: Cogan Station, PA
Me too vrg. =)91White-T wrote:You should throw a few of these together and sell em, I'm sure some people would be interested, I definitely would.
-Matt
'92 SS 5mt. All go and no show. Sold :(
'94 Audi UrS4 Modded (new project)
'96 Outback 5mt.
'07 Legacy 2.5i SE
[quote="Redlined"]
Oh... and I hope the fucker get bunked with Gunter, arrested for raping Gorillas.[/quote]
'92 SS 5mt. All go and no show. Sold :(
'94 Audi UrS4 Modded (new project)
'96 Outback 5mt.
'07 Legacy 2.5i SE
[quote="Redlined"]
Oh... and I hope the fucker get bunked with Gunter, arrested for raping Gorillas.[/quote]
-
- First Gear
- Posts: 148
- Joined: Wed Oct 16, 2002 10:50 pm
- Location: Hilliard, OH
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
Brat4by4 - I actually came up a design a while back that would pass the voltage unchanged until a threshold, and then ramp it down after that. That way the transfer function would still be continuous but you'd get a lifted fuel cut. I think I posted about it in another thread. I never did actually try implementing it though...
Part of the problem is that the stock Legacy Turbo sensor can't read more than about 14 psi of boost, so the best you could do with that sensor is raise the fuel cut to 14psi. That could be useful, but not that useful. I think it might be better ideally to set something up to use a GM 2.5-bar or 3-bar MAP sensor and incorporate a fuel cut lifter into the conditioner circuit you'd need to translate the signal.
But designing a circuit to do what this FCD does except also allow high signals through wouldn't be hard in principle.
91White-T - Hmm... I could use more money... :) Let's see how this unit does on boostjunkie's car first, after he figures out his cooling system troubles.
If you've ever used a soldering iron before you should try making it yourself... it's pretty easy, especially if you use that Radio Shack board. That's why I posted all the info on the web. The only really sensitive component on there is the IC, so you could just use a socket instead and feel pretty safe about soldering even if your skills aren't that great. Get a small project case and a few different colors of wire and it'll look really professional.
Part of the problem is that the stock Legacy Turbo sensor can't read more than about 14 psi of boost, so the best you could do with that sensor is raise the fuel cut to 14psi. That could be useful, but not that useful. I think it might be better ideally to set something up to use a GM 2.5-bar or 3-bar MAP sensor and incorporate a fuel cut lifter into the conditioner circuit you'd need to translate the signal.
But designing a circuit to do what this FCD does except also allow high signals through wouldn't be hard in principle.
91White-T - Hmm... I could use more money... :) Let's see how this unit does on boostjunkie's car first, after he figures out his cooling system troubles.
If you've ever used a soldering iron before you should try making it yourself... it's pretty easy, especially if you use that Radio Shack board. That's why I posted all the info on the web. The only really sensitive component on there is the IC, so you could just use a socket instead and feel pretty safe about soldering even if your skills aren't that great. Get a small project case and a few different colors of wire and it'll look really professional.
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
-
- Knowledgeable
- Posts: 1770
- Joined: Mon Dec 16, 2002 3:20 pm
- Location: Owings Mills, MD
- Contact:
FrmRgz2Rchz, the only two 5v sources I found on the car were the MAP power and TPS power.
Or you could take a Radio Shack 5v voltage regulator and wire it in-line with any 12v source.
Or you could take a Radio Shack 5v voltage regulator and wire it in-line with any 12v source.
[url=http://www.angelfire.com/md3/91turbolegacy/images/On_the_Lawn.jpg]1991 Legacy Turbo (RIP)[/url]
[url=http://www.angelfire.com/md3/91turbolegacy/images/Summer_Car_Wash3.jpg]2000 Celica GT-S[/url]
[url=http://www.angelfire.com/md3/91turbolegacy/images/Summer_Car_Wash3.jpg]2000 Celica GT-S[/url]
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
But you shouldn't (unless you have to because your ECU's broken) because its output won't equal the ECU's reference voltage.boostjunkie wrote:Or you could take a Radio Shack 5v voltage regulator and wire it in-line with any 12v source.
I guess I should make this clearer on that web page:
+12v should splice into an ignition-switched 12-volt source. The MAF sensor power supply is what I recommend.
+5v should splice into the pressure sensor's power supply.
Ground should splice into the pressure sensor's ground.
The pressure sensor's signal wire should be cut, with the Input being connected to the sensor side and the Output being connected to the car/ECU side.
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
-
- Knowledgeable
- Posts: 1770
- Joined: Mon Dec 16, 2002 3:20 pm
- Location: Owings Mills, MD
- Contact:
IT WORKS!!
I was able to raise boost above fuel cut, but not by much with a slipping clutch:(
SUCCESS!!
I was able to raise boost above fuel cut, but not by much with a slipping clutch:(
SUCCESS!!
[url=http://www.angelfire.com/md3/91turbolegacy/images/On_the_Lawn.jpg]1991 Legacy Turbo (RIP)[/url]
[url=http://www.angelfire.com/md3/91turbolegacy/images/Summer_Car_Wash3.jpg]2000 Celica GT-S[/url]
[url=http://www.angelfire.com/md3/91turbolegacy/images/Summer_Car_Wash3.jpg]2000 Celica GT-S[/url]
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
Yay! Experimental verification!
Slipping clutch? D'oh...
[plug type="shameless"]I've got a clutch kit listed in the parts shed[/plug]
...so are your boost-building problems and weird A/F ratios fixed?
Slipping clutch? D'oh...
[plug type="shameless"]I've got a clutch kit listed in the parts shed[/plug]
...so are your boost-building problems and weird A/F ratios fixed?
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
-
- First Gear
- Posts: 148
- Joined: Wed Oct 16, 2002 10:50 pm
- Location: Hilliard, OH
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
You should be able to get the LM224 from any electronic components retailer... Online you could order it from Digi-Key (www.digikey.com), Mouser Electronics (www.mouser.com), or Jameco Electronics (www.jameco.com)...
Some manufacturers are willing to give free samples, too... check their web sites. Texas Instruments and National Semiconductor in particular are very generous.
You might try the LM324 anyway... it's surprising how many products sold for automotive use contain components rated for 0 to 70 degrees Celsius. Almost all "Air/Fuel Ratio Gauges" for example use LM3914s which are rated for that range.
Usually overly hot temperatures are a bigger problem than overly low temperatures anyhow. And a lot of times the chips come off the same die anyway (kind of like processors' rated clock speeds). So you'll probably have no trouble with the '324.
Maybe build it with a socket, so you can swap in an LM224 if you end up having to.
Or if you want to go a little more hardcore, get an LM124, rated to military specs (-55 to +125 Celsius).
Let us know how it goes! :)
Some manufacturers are willing to give free samples, too... check their web sites. Texas Instruments and National Semiconductor in particular are very generous.
You might try the LM324 anyway... it's surprising how many products sold for automotive use contain components rated for 0 to 70 degrees Celsius. Almost all "Air/Fuel Ratio Gauges" for example use LM3914s which are rated for that range.
Usually overly hot temperatures are a bigger problem than overly low temperatures anyhow. And a lot of times the chips come off the same die anyway (kind of like processors' rated clock speeds). So you'll probably have no trouble with the '324.
Maybe build it with a socket, so you can swap in an LM224 if you end up having to.
Or if you want to go a little more hardcore, get an LM124, rated to military specs (-55 to +125 Celsius).
Let us know how it goes! :)
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
-
- Knowledgeable
- Posts: 1770
- Joined: Mon Dec 16, 2002 3:20 pm
- Location: Owings Mills, MD
- Contact:
I'm running the car at 1bar and find that I get slippage at anything below 4000rpm. I have an exedy clutch that Adam threw in with the car, but after spending money for the waterpump, I'd rather hold off a little longer before changing it.vrg3 wrote:Yay! Experimental verification!
Slipping clutch? D'oh...
[plug type="shameless"]I've got a clutch kit listed in the parts shed[/plug]
...so are your boost-building problems and weird A/F ratios fixed?
Boost comes on normally again and a/f ratios are perfect!! Waterpump solved the other problem. Now I just have to figure out how to seal the hose connection to from the intercooler hose to the bov input. Adam used a bolt-clamp on the connection that crushed the plastic connection, so the hose doesn't get clamped on tight enough to seal completely. I'm loosing boost through this connection.
[url=http://www.angelfire.com/md3/91turbolegacy/images/On_the_Lawn.jpg]1991 Legacy Turbo (RIP)[/url]
[url=http://www.angelfire.com/md3/91turbolegacy/images/Summer_Car_Wash3.jpg]2000 Celica GT-S[/url]
[url=http://www.angelfire.com/md3/91turbolegacy/images/Summer_Car_Wash3.jpg]2000 Celica GT-S[/url]
I will build this curcuit and sell it to the first group of board members near cost.
My only requirement is that it needs to be a little more universal.
Is it possible to have the unit accept upto 14 volts and add adjustability? I plan to invest a good chunk of money to build a large number if they are applicable to other vehicles.
Most likely the curcuit will be built then encased in epoxy to help protect it (from vibration and heat).
Im serious about this, I have some connections over seas that from past experience are really good.
My only requirement is that it needs to be a little more universal.
Is it possible to have the unit accept upto 14 volts and add adjustability? I plan to invest a good chunk of money to build a large number if they are applicable to other vehicles.
Most likely the curcuit will be built then encased in epoxy to help protect it (from vibration and heat).
Im serious about this, I have some connections over seas that from past experience are really good.
-
- Knowledgeable
- Posts: 9809
- Joined: Mon Jun 16, 2003 11:20 pm
- Location: Beverly, MA
You mean overseas manufacturing contract? How will you assure quality?
I did SMT process engineering as well as other stuff at a Solectron division. I think these would be rather easy to make at home. For the through-hole components, I may be able to get you an old, but functioning wave solder machine and screen printers. But I only have maybe another month before the facilities guy there is gone (whole plant moved to Malaysia).
Steve
I did SMT process engineering as well as other stuff at a Solectron division. I think these would be rather easy to make at home. For the through-hole components, I may be able to get you an old, but functioning wave solder machine and screen printers. But I only have maybe another month before the facilities guy there is gone (whole plant moved to Malaysia).
Steve
Midnight in a Perfect World on Amazon or order anywhere. The first book in a quartet chronicling the rise of a man from angry criminal to philanthropist. Midnight... is a distopic noirish novel featuring 'Duchess', a modified 1990 Subaru Legacy wagon.
"How will you assure quality?"
That is a valid question. Something I will address once I can provide a design to the guys. What would be more of an issue? DOA units, Failures during use or incorrectly built?
The investment wouldn't be too large so Im willing to take the initial risk. It can benefit so many people and I can make a little profit on the side. I hate the fact that even the cheapest FCD available is so expensive, especially when it doesnt need to be.
You should also realize that some people would rather pay another person to build it while others dont have the knowledge or tools to do it themselves.
That is a valid question. Something I will address once I can provide a design to the guys. What would be more of an issue? DOA units, Failures during use or incorrectly built?
The investment wouldn't be too large so Im willing to take the initial risk. It can benefit so many people and I can make a little profit on the side. I hate the fact that even the cheapest FCD available is so expensive, especially when it doesnt need to be.
You should also realize that some people would rather pay another person to build it while others dont have the knowledge or tools to do it themselves.
-
- Knowledgeable
- Posts: 9809
- Joined: Mon Jun 16, 2003 11:20 pm
- Location: Beverly, MA
Component quality is always an issue, but if you use reputeable suppliers your most likely failures would be incorrect manufacture; incorrect placement, missing parts, or since it's so simple, I'd say poor solder - especially if through-holes are done by hand instead of with a wave solder.
Not to discourage you, I think it's a good idea. Sell a version with a spiffy label on it for double the price to ricers.
FYI: I believe China's the cheapest for labor as Malaysians made double or more with Smart (still cheap, maybe better quality as sometimes more experienced).
Steve
Not to discourage you, I think it's a good idea. Sell a version with a spiffy label on it for double the price to ricers.
FYI: I believe China's the cheapest for labor as Malaysians made double or more with Smart (still cheap, maybe better quality as sometimes more experienced).
Steve
Midnight in a Perfect World on Amazon or order anywhere. The first book in a quartet chronicling the rise of a man from angry criminal to philanthropist. Midnight... is a distopic noirish novel featuring 'Duchess', a modified 1990 Subaru Legacy wagon.
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
ciper - Up to 14 volts? I don't understand. Almost all pressure sensors use a 5-volt reference...
As for making the limit adjustable, that's pretty easy. Just replace the 47K and 10K resistors with a 50K-ish potentiometer (the precise value isn't critical). One end connected to pin 11 of the LM224, the other end connected to the 5-volt reference, and the wiper connected to pin 10 of the LM224. Trimpots would be appropriate since they hold their position well. Then you can adjust the limit to any voltage you choose. I just picked the resistors I picked because they are easy to find, make the circuit very simple, and work well for Legacy Turbos.
As for having these fabbed overseas, that in my opinion sounds like overkill... Ordering from Digi-Key, you can pay less than $2 per unit in parts aside from any printed circuit boards. If you really want to maybe you could draw up a PCB and have it fabbed, and then spend 3 minutes and 2 dollars populating each board. Or just use the Radio Shack 2-for-$1.99 boards and spend 5 minutes and 2 dollars populating each. Like Steve says, they're pretty easy to build at home.
I absolutely agree with your statement that you hate the fact that FCDs are so expensive. That's why I wrote this up. I understand some people would rather not go through the hassle of building one themselves, but to make it worth your time it would quadruple the cost...
As for making the limit adjustable, that's pretty easy. Just replace the 47K and 10K resistors with a 50K-ish potentiometer (the precise value isn't critical). One end connected to pin 11 of the LM224, the other end connected to the 5-volt reference, and the wiper connected to pin 10 of the LM224. Trimpots would be appropriate since they hold their position well. Then you can adjust the limit to any voltage you choose. I just picked the resistors I picked because they are easy to find, make the circuit very simple, and work well for Legacy Turbos.
As for having these fabbed overseas, that in my opinion sounds like overkill... Ordering from Digi-Key, you can pay less than $2 per unit in parts aside from any printed circuit boards. If you really want to maybe you could draw up a PCB and have it fabbed, and then spend 3 minutes and 2 dollars populating each board. Or just use the Radio Shack 2-for-$1.99 boards and spend 5 minutes and 2 dollars populating each. Like Steve says, they're pretty easy to build at home.
I absolutely agree with your statement that you hate the fact that FCDs are so expensive. That's why I wrote this up. I understand some people would rather not go through the hassle of building one themselves, but to make it worth your time it would quadruple the cost...
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
Ill try to build the curcuit you mention. The reason I wanted 14 volt was to use it for other purposes in the vehicle besides fuel cut. Not absolutely necessary though.
I could send a curcuit diagram and (estimate) get 50 units for 500$. Thats complete built units in cases. I bet I could easily sell them on ebay for less than half the average price of an FCD.
If it works out well I would send you a gift in paypal as well
I could send a curcuit diagram and (estimate) get 50 units for 500$. Thats complete built units in cases. I bet I could easily sell them on ebay for less than half the average price of an FCD.
If it works out well I would send you a gift in paypal as well
-
- Vikash
- Posts: 12517
- Joined: Sun Nov 03, 2002 2:13 am
- Location: USA, OH, Cleveland (sometimes visiting DC though)
- Contact:
The tricky part is that all voltages are relative to a reference. Almost all sensors have signals between 0 and 5 volts, so that they can be relative to the ECU-regulated 5-volt supply... Some cars do use battery voltage as a reference but it's not common.ciper wrote:Ill try to build the curcuit you mention. The reason I wanted 14 volt was to use it for other purposes in the vehicle besides fuel cut. Not absolutely necessary though.
Just curious -- what other applications were you thinking of? I can't, off the top of my head, think of any other good applications for a voltage clamp on a car.
In any case, if you wanted to, you could just connect the "+5v" wire to +12v and then the potentiometer would be select a threshold as a fraction of the +12v rail instead of the +5v rail. The only catch is that the input range of the LMx24 only goes up to about 1.5 volts below the positive rail, and the output is even more limited. You could switch to a more expensive rail-to-rail op amp instead to get around that.
You probably could, yes, with the potentiometer in place. You'll need to figure out a good way to have the pot be adjustable while still having the unit in a protective case.I could send a curcuit diagram and (estimate) get 50 units for 500$. Thats complete built units in cases. I bet I could easily sell them on ebay for less than half the average price of an FCD.
The funny thing is that most commercial FCDs are probably pretty much identical to this one with a pot.
I'd definitely appreciate it if you make money off my design.If it works out well I would send you a gift in paypal as well
"Just reading vrg3's convoluted, information-packed posts made me feel better all over again." -- subyluvr2212
-
- Knowledgeable
- Posts: 9809
- Joined: Mon Jun 16, 2003 11:20 pm
- Location: Beverly, MA
You could just calibrate it with a meter and use some locktite or something to hold the adjustment screw in place. The potting could cover everything up to that point. If you're doing this on the cheap, you'll probably use a cheap hand held epoxy/potting dispenser anyway (not that an automated dispenser would be hard to program either, but the larger volumes of potting would take a while unless you get a large nozzle). Actually, there's the potting expense... OK, so I'd just pot around the stress points by the PCB, then, and this would allow for better heat disapation anyway. You'd either have to calibrate them before putting them in the case or make the case well-sealed, but accessible once assembled. How would you seal the case permanently anyway? Ultrasonic weld? I guess the easiest solution is to make them accessable.
Steve
Steve
Midnight in a Perfect World on Amazon or order anywhere. The first book in a quartet chronicling the rise of a man from angry criminal to philanthropist. Midnight... is a distopic noirish novel featuring 'Duchess', a modified 1990 Subaru Legacy wagon.