Full digital gaugecluster idea.... need input

Ok, for the longest time I have wanted to have digital gauges in my Jeep. This isnt an offroading machine (at least not yet cause Im still in college til Dec and thus have no money). This is my daily driver, and I would like to have a little input from you fellas and get a little help or advice.

I want my gauge to look like this mspaint rendering I made :



I want all six gauges on there in 7 segment and 16 segment display units. I want to use blue as shown, but I might use green. Obviously fulltime/part time wont show until active, along with "brake" and hi/lo beams.

My idea: using a basic stone aged computer running a simple program (of which I dont know what, and this is where the computer tech people come in) write code that recieves voltage data from the gauge sending units, and converts into the appropriate readout.

-convert volts from battery into....nothing actually, just display the voltage reading.
-convert voltage from oil into a range from 0-80
-convert voltage from fuel sending unit into a percentage
-take voltage from stock tach and convert into a 2 digit reading
-temp sending unit from volts to reading of 0-260 degrees
-my MJ doesnt have a mechanical speedo hook up on the NP435 tranny we are putting in it, so I will use a setup that will read driveshaft rpm a put into MPH
-write a code that uses mph x time at speed to calculate mileage.


Obviously I could use a real computer to do this, but I dont want the lag of start up time, nor the bulk. I want to use something that boots really fast. (Im not HUGE into computers/boards/electronics etc YET, but I am teaching myself.) I do have a few Dell Axim PDAs I could use as the computer, but I would rather just wire a board that does this. I'm going to try to make it plug-n-play, other than the speedo, reading off the inputs in the back of the cluster.

These are all the things I dont know how to do... I am researching to teach myself, but if anyone has ideas or inputs, then please let me know.

This will be going into a RENIX.

I know it CAN be done, but if anyone in the field of computers has input or advice, I would appreciate it.

Thanks alot guys

--Alex



EDIT: I know this is a large task, but I like challenges and this wont be done in a weekend......or any time soon really

I thought about doing this as well, but only cause I am a computer engineer and I love this type of thing. I built my own microcontroller and I have a setup to pretty much do this... only issue is actually programming the thing. It would be real easy to build something that woud allow you to do this for really pretty damn cheap. My problem is that I dont have access to the equipment to program the EPROM anymore and I prefer to work in assembly level code.

It all depends what level code you want to work in. You could always pick up a microcontroller with all the inputs and stuff set up for you and you would just have to wire and write the code. I know of a couple but I cant think of the names off the top of my head. I will look them up later though, see hwat types of inputs/outputs they have nad post them if they would work for this.

Re: Full digital cluster idea.... need input

now lets see if i can do this right?http://nordskogperformance.net/home.php?cat=69 this place has complete dig gage panel that you can build or buy complete universal panels,going to do this in my 77 f250 high boy if link don't work send email to [email protected] or call have a lot of input on this just not good typing . hope this helps? Steve n or you can call 1-602-361-1084,in phoenix az.

Re: Full digital cluster idea.... need input

stevmav said:

now lets see if i can do this right?http://nordskogperformance.net/home.php?cat=69 this place has complete dig gage panel that you can build or buy complete universal panels,going to do this in my 77 f250 high boy if link don't work send email to [email protected] or call have a lot of input on this just not good typing . hope this helps? Steve n or you can call 1-602-361-1084,in phoenix az.

Click to expand...

Are you, Steve, the one who is selling these or do you just have alot of personal information about them? The reason I considered doing my own custom cluster is price.... $500+ is not cheap to me by any means and just not worth it. <$100 and alot of time is worth it in the end to me.......my labor is free to me.

no not selling just doing on my 77 F250 4x4,just gathered a lot of info on deferant types of panels, steve n.

STACK also makes relatively cheap digi clusters (cheap, compared to F1 clusters anyway...)

it'd be GREAT to see someone here on naxja do the leg work, im sure you could sell quite a few of them if it came out nicely...

If you're a DIY type, you may also want to go to your local library and check backissues of Popular Electronics, Popular Mechanics, and perhaps Radio-Electronics. All three are geared toward the DIY guy, and I know there's been at least one "DIY Digital IP" project in there at some point in time.

Customising colours is fairly simple - you just replace the typical "red" display with whatever colour you want. However, blue LEDs are going to be fairly spendy (they're still new, y'know,) and I'm not sure if they're available in a 7-segment setup yet...

Yeah, I will look for DIY stuff in popsci and whatnot.

I would have just stuck to red or green, but according to ebay they have the blue. I haven't price checked vs another color yet but they seem reasonable enough.

http://shop.ebay.com/?_from=R40&_tr..._nkw=7+segment+blue&_sacat=See-All-Categories

I will use around 2'' for the Tach/Speedo ( I might do them on top of one another and stretch the tach into 4 digits) and then the others will be 1'' numbers. The Odometer will likely be .5 inch letters. I am thinking about using a piece of .5 inch diamond plating painted/ powder coated black for the backing. I also thought about the possibility of slapping this thing on the mill, and cutting out the lettering under all the numbers to cut cost and backing them with leds...I've got some white ones. I thought I wanted to do the 16seg, but thats just costly and dumb for something that is static like "fuel" or Temp"That way when the "low fuel" comes on I can turn the white led to red as an indicator light, same thing if I get into the red zone of temp, pressure, or volts. I think this will also help on wiring/ simplicity.

I'll cut some turn signal indicators/high beams/part time/ full time/brake emblems out too and back them with LEDs rather than writing all that crap out with the 16segments.

Stallacrew said:

Yeah, I will look for DIY stuff in popsci and whatnot.

I would have just stuck to red or green, but according to ebay they have the blue. I haven't price checked vs another color yet but they seem reasonable enough.

http://shop.ebay.com/?_from=R40&_tr..._nkw=7+segment+blue&_sacat=See-All-Categories

I will use around 2'' for the Tach/Speedo ( I might do them on top of one another and stretch the tach into 4 digits) and then the others will be 1'' numbers. The Odometer will likely be .5 inch letters. I am thinking about using a piece of .5 inch diamond plating painted/ powder coated black for the backing. I also thought about the possibility of slapping this thing on the mill, and cutting out the lettering under all the numbers to cut cost and backing them with leds...I've got some white ones. I thought I wanted to do the 16seg, but thats just costly and dumb for something that is static like "fuel" or Temp"That way when the "low fuel" comes on I can turn the white led to red as an indicator light, same thing if I get into the red zone of temp, pressure, or volts. I think this will also help on wiring/ simplicity.

I'll cut some turn signal indicators/high beams/part time/ full time/brake emblems out too and back them with LEDs rather than writing all that crap out with the 16segments.

Click to expand...

Could be fun.

If you could figure out how to drive levels (to generate polarity-reversed/AC signals,) you could use a "Tri-state" LED for your warning indicators. IIRC, they're available in Red/Yellow/Green - you simply change colours by changing polarity. Driving it with positive polarity gives one colour, negative polarity another, and AC signal generates a third.

Or, do a "pilot stack" and use a signal driver to demux into a "three-from-one" to drive a Green/Yellow/Red stack. I think you can even get pilot stacks already built, just need to be soldered down.

I've sometimes wondered if a "two-colour" (reverse polarity to change colours) 7-segment is available yet. Have it be green or yellow (or something) when everything's normal, but if the driver gets a signal out of range it turns the display red to get your attention...

Yeah I googled around and did find that red/green ones are made, but I didnt really find a price on them. It was a poorly maintained site with really no info.

Stallacrew said:

Yeah I googled around and did find that red/green ones are made, but I didnt really find a price on them. It was a poorly maintained site with really no info.

Click to expand...

You check Allied Electronics yet? They carry a lot of stuff like this - www.alliedelec.com.

Dig the sig - reminds me of the way I used to drink (less the straw.) I'd go to the Black Watch in Los Gatos every now and again, and order Kamikazes by the "pitcher" (pint glass.)

It's mainly college kids, they've got no idea how to drink properly. They'd share the pint out into shot glasses, while I was killing off pint glasses instead. Wussies.

This is an interesting thread. Not only for cluster upgrade but in general
As for development and flexebility I would go for an small PLC(Programmable Logic Controller) with EEprom option.
The PLC's are mainly for industry purpose, but would do fine in a vehicle as well.

This is a controller, not a display. To hook up an display (I.e made by your self) schould be easy and cheap.

Benefit with the PLC unit is that you have also transistor or relay output so you coould actually develope more functions to your car.

• interface a full LCD display
• own alarm, teft security
• winch control
• Light and turn signal control
• el- locking kontrol
• motor control
• transmission control
• ....limitation to your own fantacy.

It is, with other words, and extra computer, industrilized and fulle flexible
for programmin anything you need in you car.
http://electrodepot.com/al24mrd.htm
Price I guess woul be from $200 - depending of nbr og I/O needed.

Or....
Just try out some of these kits

http://www.purdyelectronics.com/products/displays/prototyping.cfm

I don't know if using a computer is the practical way to do this... but if you do, just buy a netbook instead of a stone age computer.

It would be easy if you had OBDII

MHP? :laugh2:

Stallacrew said:

I want my gauge to look like this mspaint rendering I made :

Click to expand...

winkosmosis said:

I don't know if using a computer is the practical way to do this... but if you do, just buy a netbook instead of a stone age computer.

It would be easy if you had OBDII

Click to expand...

Depends. OBD-II is essentially an dedicated embedded system with limited programmability.

A netbook is going to be fairly large, and require more power than a PLC would (or even a BASIC Stamp - I think that's just a fancified PLC anyhow...), and be larger, and be overly complex. A PLC doesn't require much operating system, doesn't require much display, and can probably be interfaced with a standard PC for easier programmability (write the programme on the PC, test it on a simulator - they're available for most common models of PLC - and up it to the "live" hardware.)

A PLC core unit can be built into something not much larger than a pack of smokes or two packs of playing cards, with the display being a separate module somewhere (could go where the OEM clock module is/was on 1996 and earlier rigs, where the old AMC "Systems Sentry" went.)

Less operating current means less heat dissipated, which opens up mounting options. Netbooks use far faster processors than would be needed for something like this - by several orders of magnitude! Slower processor means less heat. Less heat means increased mounting options. How about above the headliner (to keep it out of water when fording to the floorboards or higher,) and have the display driver cable run down the A-pillar and into the IP to drive the display to the right of the IP (as mentioned earlier?) Or, put it in an overhead console and have a flip-down for "enhanced data display," and drive a series of pilot lamps for the normal user display?

Now you're getting me thinking... And I thought I was going nuts with the idea of revamping the XJ IP to accept aircraft gages (because I like them...)

[QUOTE
Now you're getting me thinking... And I thought I was going nuts with the idea of revamping the XJ IP to accept aircraft gages (because I like them...)[/QUOTE]

Agree....why not make a package with a display to place wherever you may need it in the car. An info center with all cluster functions and i. e. your own OBD reader !...you can even map up the guidelines to help you out in a fast way.
Do not think of adding GPS, music or that kind of functions to it...than you will be lost. Keep it simple...easy intregrated and functional...would be nice with touch sceen though.....and from here...no limits at all.

Hmmm, might be worth making a kit like this for XJ's

themangeraaad said:

I thought about doing this as well, but only cause I am a computer engineer and I love this type of thing. I built my own microcontroller and I have a setup to pretty much do this... only issue is actually programming the thing. It would be real easy to build something that woud allow you to do this for really pretty damn cheap. My problem is that I dont have access to the equipment to program the EPROM anymore and I prefer to work in assembly level code.

It all depends what level code you want to work in. You could always pick up a microcontroller with all the inputs and stuff set up for you and you would just have to wire and write the code. I know of a couple but I cant think of the names off the top of my head. I will look them up later though, see hwat types of inputs/outputs they have nad post them if they would work for this.

Click to expand...

:cheers: go computer engineering!

If I had a ref for the CCD bus electrical layer protocol and data/command protocol, plus OBD II specs (haven't been able to find them anywhere... sigh) I'd be all over this. I'd probably make use of a Z80, a 62256 32kB SRAM, and a 27C256 EPROM, plus assorted 74LSxx logic and some 7 segment displays.

EDIT: I have an EPROM/EEPROM burner and an eraser that'll handle several hundred EPROMs at a time. Just need to get a win2k box running again so I can use the burner... after that, send me chips (or density/org/part number desired and money to buy) and binaries and I'll burn whatever you want. Also, for prototyping, you can always go ghetto with a Dallas Semiconductors DS1218, a lithium cell, the required size SRAM (all to form an NVSRAM of the required size), and a bunch of toggle switches and pullup resistors to toggle the data into it. Might take a while for a 32k chip though

EDIT2: if anyone can manage to find me that info, time permitting, I'll start working on a design and will publish full schematics, design methodology (such as it is, I usually keep that in my head and never write it down), parts BOMs, and PCB art.

EDIT3: also required information... input/output transmission temp. Rigging this up with a few more thermistors and ADC channels would be easy, I'm thinking a pushbutton under the temp gauge that selects engine/tranny-in/tranny-out temp readout, or a small toggle that makes it switch every few seconds. I was thinking about rigging this with a few spare coolant temp sensors and a toggle switch on my 96, but I suspect that it might confuse the ECU and/or make it run in rich mode (anyone know the temp at which the ECU decides it doesn't need to run in rich mode anymore?)

BTW there is a Bully Dog device that does all the stuff being discussed, but I don't now if it works with OBDI. I don't think the Scanguage does.

kastein:

Unfortunately there doesn't seem to be much public info on the inner workings of the CCD bus, unless you feel like obtaining the SAE paper on it: http://www.sae.org/technical/papers/880586

Apart from that, this project might server as a starting point:
http://www.kolumbus.fi/juha.niinikoski/CCD_bus/ccd_display.htm

Thanks Michael! I'll look into the project, since it describes the electrical layer well enough for me to at least capture data and try to reverse engineer it.