383LPE

I am having a little trouble trying to read my TPIS Level V chip in order to make a backup. I have read and backed up my stock chips and my LPE chip just in case there was an electrical malfunction and the programs were lost.

It could be a bad connection from the carrier to my emp reader, or a read protection built into the chip. This is a 93-94 chip and its seems unlikely there is a read protection on it. Although there is plenty of silicone and gold paint and there is no way to define the chip device (although it is UV erasable). I am selecting the 27c256 and getting some ff's with some data sprinkled throughout. definately not a complete bin file.

Also, in the past I have had the same experience as TJWONG, reading "custom" programs that actually reflect the stock program other than fan on-off temp changes etc.. I am hoping to see developed VE tables etc.. for what TPIS gets for these chips.

Any help is appreciated.

tjwong

Sometimes unscrupulous chip programmers will clip one of the pins on the prom in a attempt to thwart those who would want to read their chips. Check and make sure that all the pins are intact on the prom. Is this prom on an adapter board or is it a complete memcal like a GM memcal? I have seen this done by TPIS in the past on the big blue memcals, as its easy to hide a cliped pin on the prom under the blue cover.

Then some promgrammers will put some "trash" into the file so that TC or another ECM program will have a hard time reading it. I have sucessfully "hacked" most of these attempts that keep people from reading the files. If that is the case feel free to email it to me and I will see if I can hack it for you.

383LPE

This sounds pretty weird, but they siliconed around the chip on the memcal, applied the blue cover while wet, put a smiley face sticker on the UV window and spraypainted the whole thing GOLD. I guess the gold paint is to make you feel warm and fuzzy about the 600 bucks you paid for the chip. This chip was installed in 1994-95.


[Modified by 383LPE, 9:48 PM 5/6/2004]

tjwong

If I remember right they would clip pin 1 that would make it so that no all addresses are accessed during the read. That would explain a lot of FFs and some data.

dgoodhue

I was able to read my TPIS chip, but they changed the code so that I couldn't access the info (they changed the defination file designation). Mine was an older chip though.

tjwong

You can send me the file and maybe I can hack it.

383LPE

just got it cleaned up. In fact they did cut one of the eprom pins.

383LPE

It will open in 8D, problem solved.


[Modified by 383LPE, 9:31 PM 5/6/2004]

1MoorTym

I'm not an expert on this, but I have a couple of comments.

1. Pin 1 on a 27c256 is the programming voltage pin. Cutting this pin would only affect the ability to re-burn it once it had been erased.

2. Any pins that are cut in attempt to stop you from reading the program such as an address or data line would also stop your ecm from reading the program, rending the chip worthless.

3. Are you sure that chip is a TPIS piece? The silicon, paint and smiley face thing sounds odd. I just bought a chip from them a couple of months ago and its a professionally done stock looking memcal. Oh, and their level V chip are $575

383LPE

1. Actually its pin 28 that IS cut. I had to bypass it to get a good read.

2. Obviously not, since the car is running off the chip with a cut pin

3. Yes, I am sure its from TPIS, and was originally a level 4 then sent back for an upgrade to a level V. As I write, I am doing a comparison between the TPIS 383 chip, LPE 383 chip and a stock GM 91 chip. The TPIS chip thus far has been an exact copy of the stock bin (fan on-off has been changed plus some minor needless changes to protect themselves), and no where near as developed as the LPE chip. Even at 575, you basically bought a stock memcal.

I will report with more as the study continues.

tjwong

If you got a good read from the chip, and TC tells you that it requires a different definition to read it then they skewed some information on the prom hex file. If you send it to me I can fix that as well or at leas attempt to do it. Normally I can fix 99% of them so called "protected" files.

Typically on a 27C256 or a 27C512 chip pin 28 is known as the Vcc leg. 5 volts has to be applied there and to pin 14 which is DC ground in order to read the chip. 5 volts is only present at Vcc pin 28 when a read is in progress. So this means that if the sly little guy like Myron or his programmer decides to clip the lead, you can't read it. Hence the reason for his attempt to "pot" the chip with silicone glue just to discourage you some more in your attempts to read it Sneaky like chumps aren't they



[Modified by tjwong, 7:47 PM 5/6/2004]

1MoorTym

Pin 28 on a 27c256 is VCC - chip voltage - the chip would not work with that pin cut. Are you sure its a 27c256 ??

Not trying to give you a hard time, but cutting address or data lines affect both your chip reader and the ecm's ability to read a chip. Think about it, the chip doesn't know where its plugged into right? So if the ecm can read all the data using the address and data lines how come the chip reader can't?

I'm also not sure how they can skew the data. Isn't this just 1's and 0's located at specific locations where the ecm expects to find it. Thats why you need the right definition files for the program your reading to organize the data correctly. Doesn't the ecm require data values at known locations? If they move the data arounbd on the chip won't that screw up the ecm's ability to read the right data values from the locations it expects to find it at?

Like I said, not an expert...

383LPE

You are absolutely right.

BTW, on my 90 when we changed the rear gears it seems we changed a gear drive to adjust the speedo. I assume it is the same for the 91. the TPIS chip says on the box that it was spec'd for a 3.73 rear gear. Is there a bin parameter to correct for rear gears?

383LPE

The inital skewed data was me not knowing they had cut a leg off the chip do to the glued on cover etc. The def file error and funky chip reads happened during my learning curve holding a paper clip to jump the chip to the memcal. I had some bad reads when my technique was not just right.

I agree with you and do not think they use programmed chip protection, only physical chip protection. This is the first time in 4 or 5 years of reading chips that I have seen this.

Here is a pic of the memcal with the chip portion of the cover peeled off and my paperclip jumper. Pretty creative!




[Modified by 383LPE, 10:10 PM 5/6/2004]

tjwong

I am not an expert at memory devices either, I know enough to be real dangerous What I meant by skewing the number and I guess I should have choose my words better. Because they are not really skewing it. In reality they put in some trash so that Tuner Cat and other ECM programing software won't interpet the information correctly becuase their ECM definitions won't interpet it correctly. The information is all there but the definition file will not recognize it. I use HEX Edit to locate this "trash" in the file and repair it so that its once again readable. I do the same thing with encrypted LT1 Edit files from certain tuners that choose to make their files unrecognizable to Tuner Cat or LT1 Edit users.

The Vcc leg is only used during the reading of the chip. As far as I can tell it is not used during normal running in an ECM. I have checked this on a test board during simulations and there is no voltage present on that leg during normal run mode. The output enable (E) leg is pin 20 has voltage as well as Vpp pin 22 during normal run modes.

Skippy Stone

My 2 cents. I had a TPIS chip and there were not that many changes from stock. It inspired me to do my own programming.

Happy 1s and 0s....

dgoodhue

I sold the chip 2 years ago, and didn't bother keeping it after ahrd drive problems. I had talked to some guys on the 3rdgen.org PROM burning forum about how TPIS changes the file, the definition file and checksum would need to be changed. It was too much effort for baseline.

tjwong

There is a constant called "road speed constant" in the ecm constants table. You just tell it how many pulses per mile. This is the constant that lets the ecm calculate road speed. It really doesn't have anything to do with the speedo as its an internal calc for the ecm to calculate speed for its own use.

1MoorTym

Take a look at this and let me know what you think afterwards.

http://www.vgwiz.com/files/27c256.pdf

tjwong

Yes I read that and its basically the same stuff from Texas instruments for their 27C256 and 512k chips that are used in the truck ECMs. The real test is to measure the voltage in the ECM with the truck running. Thats what I did some time ago. Check it out, that leg is not powered up during ECM operation. At least not in the truck ECMs that I was messing with.


[Modified by tjwong, 7:26 PM 5/7/2004]