(aka More microcontroller madness from me again :) )

Looking at the price and quality of boost gauges I decided that I would make my own for the sprintex supercharger kit (since it already uses a 2 bar MAP sensor, why not?)

It measures the voltage across the map sensor (which senses boost and vacuum signal) and displays it numerically (to 1 dp) on an LCD screen, along with a fancy bargraph display.

Full scale deviation on boost is 8 psi - it will measure up to 14.6 psi numerically but the scale will show over-range (all segments blacked out). Vacuum will measure ~28 InHg - no FSD shown (as scaled to 30 InHg FSD).

The LCD display is quite readable in sunlight, and is backlit at night. It lives above the centre dash vents (where the fuel and battery warning lights are on an AWD - you hack up a clock from a TE to do this)

Sample Displays:

On boost, display sweeps from Left to Right

Boost 5.5 Psi
|||||||

On Vacuum, display sweeps from Right to Left

Vacuum 10.1 InHg
................||||||

Photos tomorrow (battery is dead on the camera AGAIN).

I can probably provide a pre-programmed chip if anyone wants to build one.

EDIT: Stupid forum trims spaces so I will have to use fullstops to show spaces on display :P

I love your work Heath. Always an interesting read.

sounds good, picsorban :P

mmm, are you interested in making more than one of these?

or writing up some instructions on how to make one?

I've been looking at that space in my TJ II and thinking something needs to go in there.

cheers,
.wook

Not a problem - few minor bugs I'm working out, one of which is the fact that 5V regulators don't produce a nice steady 5V - they fluctuate between 4.5 and 4.98V as input voltages change which is a PITA when you are trying to reference it for an ADC (!)

I was also going to improve it a bit - goodies like temperature sensing (using LM35's) for intake air temp etc. (0-150 degrees), welcome screen (like displaying Mitsubishi AWD Supercharged for 3 seconds on startup), single temperature display (selectable between IAT, Water temp, oil temp, trans temp etc etc. Or display up to 8 items at once.

Not entirely happy with my implementation of the bargraph either. Its a bit clunky down low, I may need to have another look at the maths behind it. (add some averaging as well).

Cheers
Heath Young

can you make it say " Hello Michael "

haha

maybe that was funnier in my head ??

can you make it say " Hello Michael "

haha

maybe that was funnier in my head ??

it sounds good in my head too but reduces ur market when your trying to sell it as you can onmly sell it to pplz called michael lol

it sounds good in my head too but reduces ur market when your trying to sell it as you can onmly sell it to pplz called michael lol

or people that like Knight Rider... see, it was much funnier in my head. :D

Could this be adapted for a turbo? I have yet to find a gauge I want to put in the Saab and the standard boost gauge (in reality a torque request gauge) isn't very useful. Also the IAT and other features sound fantastic.

i had a hidden boost gauge on my set up, under the cup holders! only used it when i felt the car was going slow and if it showed 0 psi it was cos the little vac hose kept splitting, cheap crappy rubber!

in reality you really dont need one as it kind of operates like this with a sprintex kit.....

idle----- vac
cruise---vac
press the accelerator, engine under even the smallest bit of load and over 2200rpm 6.5psi

if your gauge varies between 2500rpm and 6000rpm when under load at wot its broken or dosent work! (you will notice small variations on a digi gauge but on an analogue one you'll notice nothing.... if you do, you have a crappy boost gauge.. i have the dyno printout to prove the stabe constant boost that is practically a flat line!)

Interesting - I would have thought that there would have been variation based on RPM...

The IAT sensors will be useful, as would EGT to keep an eye on things...

So what would this setup cost and how hard would it be to wire in? Any pics yet?

'Tis pretty cheap. Not too much to wire up either, but I hit another snag - 40 pin IC's dont like being de/re-socketed often and has broken off a couple of pins, one being the program pin :(

So looks like it is down to Jaycar to buy another one - grrr...

Sign me up for one please.

What sort of regulator circuit are you using? A basic regulator should perform better than that.

Ken

7805 Decoupled with 2 X 0.1uf mylar and a 1200uf resovoir cap. This is by varying the input voltages across 9-15V... I'm thinking that the 7805 may be a bit crook (purchased new from Jaycar about a week ago - unlikely) but hmmm...

Interesting - I would have thought that there would have been variation based on RPM...

The IAT sensors will be useful, as would EGT to keep an eye on things...


It does, it starts at 6psi and ends at 8psi.

It does, it starts at 6psi and ends at 8psi.

in your car, mine was a constant as soon as it peaked boost it was flat till redline...

maybe the 8psi you have is due to restriction... with better flowing heads your installing the boost will drop! and power will go up!

The spec sheets for the 7805 should be readily available on the web. From my recollection, it sounds like you have the recommended caps in the circuit. I'll dig around for some more circuit notes.

Ken

Any movement on this?

My expensive gal ep programmer has died, and i'm looking for excuses for work to buy me one at the moment :)

Ahh, sorry to hear that.

LM7805 specs are between 4.8V - 5.2V @ 25deg (4% tolerance) so there could be something wrong with your regulator. The caps you have are incorrect though.... 330nF on the input and 100nF on the output. Neither need to be polarised. Using a COG/NPO Dielectric Multi Layer Ceramic cap will give you the best stability. As long as the input voltage remains at least 7V you will be fine (2V greater than the required Vo).

If you want anything more accurate than that, you need to use a precision voltage reference chip or a LC network on the ADC ref pin to try to stabilise it a bit (10uH in series with a 100nF cap as close as possible to the ADCref pin). Even better, if you can use an internal Vref on the chip itself, I know the Atmel AVR series have this, but its only 2.56V.

Hope that helps...

Cheers
Pete

LM7805 specs are between 4.8V - 5.2V @ 25deg (4% tolerance) so there could be something wrong with your regulator. The caps you have are incorrect though.... 330nF on the input and 100nF on the output. Neither need to be polarised. Using a COG/NPO Dielectric Multi Layer Ceramic cap will give you the best stability. As long as the input voltage remains at least 7V you will be fine (2V greater than the required Vo).

If you want anything more accurate than that, you need to use a precision voltage reference chip or a LC network on the ADC ref pin to try to stabilise it a bit (10uH in series with a 100nF cap as close as possible to the ADCref pin). Even better, if you can use an internal Vref on the chip itself, I know the Atmel AVR series have this, but its only 2.56V.

Hope that helps...

Cheers
Pete

lol inductor in series with a capacitor? Thought capacitors are open circuits eg. don't conduct current.

lol inductor in series with a capacitor? Thought capacitors are open circuits eg. don't conduct current.

LOL sorry I worded that very badly... only noticed it when I read it for the 2nd time then! Made sense the 1st time :P
No DO NOT put a cap and inductor in series together... not a good idea LOL

I'll try again with a bit more care...
Put a 10uH inductor in series with VCC and the Vref pin on the IC, then between the inductor and the Vref pin put a 100nF decoupling cap. As close to the Vref pin as possible.

Hopefully that makes more sense. :D

I would be very interested in one of those. Have you thought of using an LED bar from a KJ Verada? It goes in the same spot and still has the battery/oil lights (and who really needs the seat belt indicator). They have an LED number display that they use for the Trip Computer. Could maybe tap into those display circuits AND it would look like it meant to be there, well, because it is from there.

no offence, but wouldn't it be easier to go and spend $50 on a drift boost gauge? they are reasonably accurate.

Yeah probably, but those dont have things like selectable IAT pre and post blower, EGT, trans temp and oil temp. :)

If I can find 5 minutes to work on it again (yeah right) I should have an update soon.

I would be very interested in one of those. Have you thought of using an LED bar from a KJ Verada? It goes in the same spot and still has the battery/oil lights (and who really needs the seat belt indicator). They have an LED number display that they use for the Trip Computer. Could maybe tap into those display circuits AND it would look like it meant to be there, well, because it is from there.
Hey HeathYoung, how are we going with this project?

I have recently tried hooking up a laptop to the SMT6 ECU, but the damn thing wouldn't communicate (not sure if I was doing it right or the COM1 port on the laptop was dead). So, if your thing does the trick, well cool. I am trying to get a feel for the IAT, on and off boost. So, I can guage how effective the water injection is, once I install it, using pre-water injection and post-water injection IATs. If all looks good then ..... great, else I may need to upsize the spray nozzles.

Its a work in progress still, been stupidly busy with boring stuff like work and family.

The car is also giving me the ****s at the moment with fuel consumption. I did find a very nice white led backlit LCD, I'll get back to this soon, work is a little less stressful at the moment.

IAT isn't that hard to do with simple sensors, Jaycar have a dash-mountable max-min thermometer, it responds quickly, has a smallish thermal mass and can easily be mounted in the intake pipe and if you drill a small hole, can place it post-supercharger as well.

If I could be bothered with the conversion tables, I would also knock up one that read directly off the SMT6 thermister. Hmmm... Black wire on IAT sensor goes to ground, white wire goes to +5V via a 2.2K resistor, forming a resistive divider that could be sampled via a buffer to determine the intake temperature.

Ditto with boost, can read the map sensor directly. I'd have to make a lookup table though for the thermister, stupid non-linear sensor.

Cheers
Heath Young

Would this digital boost gauge be fitable in a small enclosure which could then sit atop the steering column? I would definitely be interested in something like this :)

Yep, sure. There are some very small backlit 16X2 displays around. Smallst I can get my hands on at the moment cheaply is a display area of 36W X 10H. You might not be able to read it though :)

I also thought about using a custom charactor set to do a HUD but there is little time available at the moment. It was going to sit in the drivers vent with a small piece of tinting for the reflector. I built a boost gauge this way, looks very nice. Used a set of high intensity 3mm leds in a 240 degree circle, dimmed at night (actually used an ambient light sensor which adjusted the brightness) Probably looked a bit weird to oncoming traffic though :doubt:

in your car, mine was a constant as soon as it peaked boost it was flat till redline...

maybe the 8psi you have is due to restriction... with better flowing heads your installing the boost will drop! and power will go up!

On the dyno mine was the same, it would show 5psi@2000rpm, peak @7psi around 2500rpm, and stay constant till the limiter.