hally

Hi well for my Chrismas present i ordered myself 'The Pocket Programmer 2' to clean up my fuel maps.
The EPROMS that i recieved are numbered HN27C64G-15. (not what i ordered which were 27C64A http://www.jdr.com/interact/item.asp?itemno=27C64A-150>
)
Looking these 27C64G-15 EPROMS up i found that they have a programming voltage of 21v. I was able to successfully burn and verify (read back) this chip. However i am concerned that they are not compatible with the DME board...
Could anyone advise? Also what specific EPROMS have people had success with
cheers

Bri Bro

The programming voltage is determined by the programmer after you tell it what type of IC you are inserting. After it is programmed, the readback will be the same, so it will work in the DME.

How do you know were the fuel maps are stored on the IC and how do you know what to change them to?

awilson40

I concur with Brian. I have used 12.5v, 21v and 25v chips in my car.
All worked fine....just dont use the higher voltage programming a chip requiring a
lower voltage or you will fry the chip.

hally

Hey Brian, just soldering in that circuit of yours for the boost signal logging (klr pin 5) -> tech edge at the moment.
Well when i select 27C64 in the programmer it comes up with 12.5 volts which fails to program. There is no specific listing in the pocket programmer device types for a 27C64G so i looked it up on the web and it had 21 volts. I tried that and it successfully programmed.
OK some test results..
It is initially working when i install in the car (enough to get some good logging done and see the changes in the A/F, i using the montronic editor by Mark SkarShinski). After i switch off the car and leave it for a few hours / over night, the car doesn't fire at all when attempting to start. With the last EPROM i tried today, the car failed to start straight off.
Not sure if my problem is the EPROM type or my programming / erasing technique
After removing the EPROM from DME the pocket programmer can still read it fine.
I have tried erasing the EPROMs for different lengths of time (4min, 6min, 8min, 12min)
cheers

hally

thanks for that info awilson... so the EPROM should be ok, something else is amiss..

awilson40

Yep, that doesnt sound like an eprom issue. How long does it have to sit before failing to start???
Thats a hard problem to track down. I would check the temp sensors at the DME, both sensors.
Check your speed and ref sensors again at the DME. Also swap out your DME relay.

hally

The car works fine whenever i pop the guru chip back in, and one of my 4 programmed chips has held up overnight which is encouraging.

Question - is the EPROM read only on DME power up (eg into some memory buffer), or are the maps actually continuously read out of the EPROM when the engine is running? I am thinking on power up as once the engine starts i have had no issues.

Bri Bro

I have tried erasing the EPROMs for different lengths of time (4min, 6min, 8min, 12min)
cheers

This is a hard one to answer, it depends on the IC and the type of UV bulb used. Bottom line, erase for 20-30 minutes and make sure your bulb has been replace in the last decade. This type of IC (EPROM) was replaced by EEPROM and flash memory that don't need a UV light to remove the charge from the cells... so the bulbs tend to be a little old.

As far a s maps are concerned, here is my guess. The 8751uP used in the DME didn't have a bunch of RAM memory inside of it. I would bet they use RAM for variables and read the maps from EPROM. Why store a number in RAM that doesn't need to be modified? The map values are just plugged into an equation to determine the proper duty cycle and ignition timing.

hally

Things are looking better, i discovered a small bug with the Montronic Editor ( http://www.mskar.org/dme_download.html ) where the max RPM gets modified each time you save the binary. Think it is doing an extra hex coversion or something along those lines. This modifies the first byte of the bin, easily fixed, but watch out for it. Doing a file compare against stock & guru chips highlighed this issue. Thought initially it might have been a checksum or something legitimate. Since fixing this issue (i just hex edit the bin after a save) i have had great success with the HN27C64G-15 EPROMs i've programmed, WOT fuel curve is looking the goods. Being up home in the country for christmas was handy for churning out the 4th gear pulls to redline
Brian, these EPROMs seem to be happy at 12mins of erasing, i'll try longer if i have any more failures. I think i fried one trying it at 25volts
thanks for info everyone.

Bri Bro

Thanks for the link. What are the PKCS #7 Certificates files for? Not much in the readme file.

Here is what I dug up in the spec sheet for the erasing this series of EPROM.

Need a UV bulb of 2537A, 12,000uW/cm**2 and it should be placed within an inch of the window for a minimum of 20 minutes.

I have erased ICs in less then 10 minutes and they will appear to be blank, the problem is that a cell or two might be close to the edge and flip its state a day/week/month later. This is can cause some really bad things to happen.

hally

The PKCS files are the mapping spec files that tell the editor where in the bin file to look for the maps. After u load a bin, you then select a mapping spec file. The '951 DME origional' addresses the 1st out of the 3 sets of maps on the EPROM which is what my car runs off. Too easy (:
You can also create your own spec files, eg to edit one of the other maps on the 951 eprom.
OK i'll bake for 24 mins. cheers