Turbo Dodge Forums banner

1 - 20 of 109 Posts

·
Registered
Joined
·
694 Posts
Discussion Starter #1
Does anybody know which EEPROM chips were used in the late '80s to early '90s digital clusters?

These are the markings on what I am trying to ID:

GI (manufacturer)
2208
8425
14-Pin

Harris nSemi
4374040 (Chry P/N)
H 9201
14-PIN

Thank you
 

·
Administrator
Joined
·
3,969 Posts
Those numbers don't come across me as EPROM part numbers, but I plugged them in at work here and checked my catalogs and I can't find a match for any kind of IC.

It is possible, unlikely, but possible that the IC's were one time - spiecal made for the clusters only. Which would be bad because you would never find a new example.

I would like to know if anybody turns something up, because I have been buying EPROM's, IC's and microswitches for TD's like crazy recently (dirt cheap through work) to have a supply, everything is being factory discontinued:bang head
 

·
Registered
Joined
·
654 Posts
May I ask why you are looking for replacement ICs? I'm glad to learn that we have a few EE/CE people on here!
 

·
Registered
Joined
·
694 Posts
Discussion Starter #4
I am aware that the numbers that I listed are not standard part numbers. They are the propreitary Chtysler part numbers.

I am looking for id's on these chips because I would like to know the format that Chrysler stores the odometer inforation in, for these years. (Essentally for my own EE edification). In the spring, I will be doing a bit of work on my '93 Spirit. One item on the list is to install one of my digital clusters. I would rather pull the EEPROM, reflash it, than hook a pulse generator up to the speedometer input to roll it forward.

(For those who have seen some of my other posts, the SBEC-II schematic project is ongoing: One PCB layer left to delainate and digitize, then the fun really begins.)
 

·
Registered
Joined
·
694 Posts
Discussion Starter #5
For those who are interested:

There are two power leads that got to the digital clusters. One is constant power, the other is switched (ignition) power. On the non CCD clusters, the speedometer signal is accumulated by the MPU on the cluster logic board, as opposed to the cluster's display board. When the switched power is lost, an erase/write routine is initiated and the speedometer pulse sum is written to the EEPROM. On power-up, this data is loaded into one of the cluster's MPU registers and the speedometer pulses are accumulated and summed to it. The cluster's MPU has two divide constants: one for the metric (kilometer) display, the other for the English (mile) display. By toggling between the two constants (M/KM button), the cluster's MPU knows which to divide the accumulated speedometer pulses by, before sending the data to the display driver circuitry.

By powering a digital cluster on the bench, one can pulse the speedometer input and advance the mileage, as well as drive the speedometer display. If the constant power is removed, before the switched power, the mileage increase will not be stored, and the previous mileage will be displayed upon the next power-up.
 

·
Registered
Joined
·
654 Posts
I haven't seen any of your other posts yet. Are you attempting to reverse engineer the SBEC-II in order to build one from scratch with some critical improvements? Do you have a link?
 

·
Registered
Joined
·
694 Posts
Discussion Starter #7
ZoskalonMPI --> I am reverse engineering the SBEC-II to get a better understanding of how it functions, as well as to dispell the "black-box" mentality that is pervasive in the auto-mod world. Mainly, I am looking to see the hardware interactions of the SBEC-II's calibration (program) code. I am going to be doing a TBI to MPFI conversion on my 2.5, 1993 Spirit. I would like to figure out how to add the coilpack code into the calibration file and eliminate the spark jump in the distributor. The wiring aspect is not an issue: I have already added power locks, power windows, Bendix-6 ABS (with 4-wheel disk conversion), the premium overhead cluster, traveller, power seats, auto climate control, and a few other amenities. The electronic cluster, EVIC (Electronic Vehicle Information Center) and 22-function EVA (Electronic Voice Alert), as well as new paint, will be comming this spring.
 

·
Registered
Joined
·
654 Posts
Where is all this work shown??? I would love to see everything you've done to this car! I don't recall ever seeing EVIC on a Spirit before, only the 2 button travelers. What do you mean by "premium overhead cluster", is it just the temp/compass console? What year ATC did you use? Info and pics PLEASE! Lol.
 

·
Administrator
Joined
·
3,969 Posts
Mangelhaft, stupid question, but the information you gave from the first post is of the IC's themself? I just wanted to make sure. If that is all the information you have to go by it may be very hard to find a datasheet on one of those without harrassing Chrysler every hour day after day.. or back probe the IC's in the cluster with the dash torn apart while somebody drives:First TD:

If you do come across a datasheet or even functional schematic of the whole digital dash post it up, it might help to revive the dying breed.

I have yet to dig into my two digital clusters for my daytona, they work but have some funny speedo bugs under vibration I have to work out.
 

·
Registered
Joined
·
694 Posts
Discussion Starter #10
The numbers that I listed in my previous post are the numbers that are stamped on the chip. Chrysler, as well as other manufacturers, had their ICs custom marked. An example is the chip marked SC80575VFN 4632459 which is a Motorola 68HC11A1 used in the SBEC-II. SC80575VFN is the motorola reference part number for the chip with the custom Chrysler ROM mask and 4632459 is the Chrysler part number. Another example is the Interstil SD68HC68S1 which has the Chrysler part number 4374040. It is used to decode the CCD databus for the chrysler traveler modules. This is usually for preprogramed MPUs, but also applies to other chips that were replaceable, or serviced by Chrysler and/or its authorized service providers. Chrysler has part numbers for the odometer EEPROMs because they could be replaced or reflashed when a cluster was replaced. This was before the time when the odometer info was stored in the BCM or ECU.
ZoskalonMPI --> As for pictures, I have not taken any, yet. I am still in the process of reupstering the interrior and prepping for paint this spring. The AA bodies never came with an EVA, EVIC, nor auto climate control. The overhead cluster is the one with the compass and temperature. There are provisions on the board for two more buttons and some unoccupied pads for additional electronics. I have yet to figure this out. The overhead clusters with the tripodometer pulled their data off of the ccd bus, and was supplied by the BCM. I added the two button traveler, but would like to add the twelve button traveler, once I figure-out the injector pulse feed, and finish the dashboard transplant. As for the climate control, I believe that it was from an '89 New Yorker. I replaced the sixteen-year-old heater core (Christmas eve 2009) and figured that since the dash was out, I might as well add this.

Why all this work? I bought my 1993 Spirit, in 1999, when I was stationed in Texas. It's body is solid and in great shape. It has only seen a few Michigan winters. I am getting c. 32 MPG, average. It is very reliable. And last, but not least: I am an engineer, so why not?

For anybody that is looking at doing some of these "upgrades", do not be intimidated by the wiring. Check with your local library for the Mitchell's Manuals. Their wiring diagrams are superior to AllData. Remember to get the connectors. Solder and heat shrink the splices: DO NOT USE CRIMP CONNECTORS! Take your time. Be patient. Ask questions. Have fun.

(Now is it worth adding the cell phone visor and associated electronics" just because...)
 

·
Registered
Joined
·
694 Posts
Discussion Starter #11
I for got to add that I am sure that there is a standard EPROM that is compatable with the Chry P/Ns. If all else fails, I can trace the EEPROM datalines and construct the pin out, and use trial-and-error to acheive a coherent ROM dump.
 

·
Registered
Joined
·
97 Posts
I for got to add that I am sure that there is a standard EPROM that is compatable with the Chry P/Ns. If all else fails, I can trace the EEPROM datalines and construct the pin out, and use trial-and-error to acheive a coherent ROM dump.
I have a 1985 Laser XE cluster on my benchtop right now with the ODO chip signals on the logic analyzer. You have to be careful as there are two negative power supplies to the chip. There is a -5 which probably would not kill my test equipment, but the apparent programming voltage(always present) is -30V!

I started by probing each pin with a volt meter to try and determine the power supplies. There were several pins that are always low, so I'm not sure which of those are the actual ground power pin. One pin seemed to be steady +5 V, so I'm assuming that's the main Vdd. The negative programming voltage is curious, most EEPROMs I've seen use +27V or so for programming. -5 V is not unusual for older chips. Perhaps this is an NMOS device and not full CMOS?

I could only identify five signals that actually toggled during the apparent read cycles(On power up) or write cycles (On 12V ign powerdown with 12V batt still powered). This suggests that the device is in some manner serial instead of parallel which will make decoding a bear.

With the apparent writes, the signaling made some sense with one of the signals appearing to be a clock or strobe of some sort. The period was 80uS which looks right, but I haven't figured out which signal(s) are data, chip select, rd/wr yet. Also, there seem to be a very large number of repeated bursts for the shutdown sequence compared to the startup read sequence which looks like one single burst after some sort of setup sequence.

Reads look totally different, where the writes look slow and synchronous, the reads look completely asynchronous, and the signal timing looks impossible. There are several short (10-20ns) pulses and delays that can't possibly be actual chip timings as there is no way 1984 automotive technology was running at 50-100MHz. What looks like synchronous 10-20ns setup/hold timings must be the actual propagation delays of these old devices. I clearly have more work to do here...
 

·
Registered
Joined
·
97 Posts
I've solved part of the Odometer Chip Mystery! This part is a renumbered GI ER-1451 700-bit serial EAROM(Electrically Alterable ROM)! I just picked up a handful of new blanks on eBay. With the pinout and datasheet for the part handy, I could easily setup my logic analyzer to take ROM dumps. On power up, the dash will dump the contents of the ROM which contains the regular ODO and the trip ODO values. On ignition powerdown, it erases and writes the new value. Of course, that means at around 10K ignition cycles, the chip is going to go bad and give you the ------ of death. Anyhoo, I still need to write a perl script to turn the logic analyzer listings and turn them into a readable ROM dump (Basically de-serializing with software). Then I can correlate the mileage on the ODO with the contents of the ROM to figure out how to burn a ROM with any arbitrary mileage in the ROM. I'm not doing this so that people can roll back their ODO's, it's so people can swap a cluster or fix a broken ODO, or if they restore their car and want it to say 0 miles. You can always check the box on the title that says "not actual mileage" and no one is going to believe that a falling apart rust bucket actually has 400 miles on it :p At this point in the life of these 80's and early 90s cars, the number on the clock means very little compared to the obvious condition of the car as determined by inspection...
 

·
R.I.P Dennis Jarvis
Joined
·
5,369 Posts
Here is a link to a website that has some source code for our ECU's. Be a good starting point. At least with the code you can figure out your input/output, and which one's are analog.

I don't know of a good decompiler for the CPU. Depending on the CPU you can use ASMFLOW to flow chart it.

Later Dennis
 

·
Registered
Joined
·
718 Posts
If any of you figure out how to get these chips to work let me know please. I need one done for a 1986 Chrysler Executive Limo (Stretched Lebaron) at some point. I have the current ------- thing of death on mr digital dash.
 

·
Registered
Joined
·
694 Posts
Discussion Starter #16
Great information. Here is my website and what I have done with the traveler: Mangelhaft's Chrysler Info I believe that the traveler uses the Motorolla 6805 MPU. I do not have the hardware to do a dump, though.

I am almost done with the SBEC-II. I need to clean-up the schematics. After that, I will start on the digital gauge clusters.
 

·
Registered
Joined
·
97 Posts
Here is a link to a website that has some source code for our ECU's. Be a good starting point. At least with the code you can figure out your input/output, and which one's are analog.

I don't know of a good decompiler for the CPU. Depending on the CPU you can use ASMFLOW to flow chart it.

Later Dennis
Hi Dennis,

Can you repost the link? It's not showing up. I also have an ECU re-engineering project planned(for Pre-'87 separate Logic Module units)
 

·
Registered
Joined
·
97 Posts
If any of you figure out how to get these chips to work let me know please. I need one done for a 1986 Chrysler Executive Limo (Stretched Lebaron) at some point. I have the current ------- thing of death on mr digital dash.
I'm getting close, but It looks like I'm going to have to build a burner for the ER-1400 chip as it uses some strange power supply (-30V). Its a pretty simple burner with a schematic and software online, but it may take me a few months to get the time to construct it. I ordered some blank chips on eBay and I plan to offer ODO chip replacement for a nominal fee to cover the parts cost and recoup the cost of building the programmer as a service to us Digi-Dash enthusiasts. Most likely your chip is dead as after 10,000 ignition cycles or so the chips cannot be programmed anymore as they electrically "wear out"...
 

·
Registered
Joined
·
10 Posts
I currently have the ------- display of death on my 85 laser XT. It failed in 1999. I have probably driven the car 50,000+ miles since then. If I replace this chip you are talking about, should I assume the odometer will work again? Do you know where I can find a repacement chip? Also, where is the chip located? I assume the car's computer no longer has any idea how many miles it should read, nor do I. Any help you could send would be great. I posted some pictures of the car if you would like to take a look. Thanks.
 
1 - 20 of 109 Posts
Top