MIL light is on, bad battery in ECU Ram chip
Discussion
Just FYI, if anyone wants to replace the Dallas chip. I used this supplier https://www.mouser.co.uk/ProductDetail/Maxim-Integ...
Also supplied by http://uk.farnell.com/maxim-integrated-products/ds...
Both are 70ns parts which is overkill for the MBE ECU but all varients are a similar price which is ~half the price at TVR outlets.
Also supplied by http://uk.farnell.com/maxim-integrated-products/ds...
Both are 70ns parts which is overkill for the MBE ECU but all varients are a similar price which is ~half the price at TVR outlets.
Noblebenn said:
...
Fast forward a year and it happened again. Started to happen every 1 in 10 starts until it became 1 in 5.
It would seems that something is causing the chip to deteriorate slowly and when the car doesn't fire up first time the adaptive maps are lost.
....
It's hard to believe the chip is deteriorating. Inside the "chip" is only memory, a battery and some logic to switch between battery or external supply. In other words, the only thing that can deteriorate is the battery. Fast forward a year and it happened again. Started to happen every 1 in 10 starts until it became 1 in 5.
It would seems that something is causing the chip to deteriorate slowly and when the car doesn't fire up first time the adaptive maps are lost.
....
When it happens, are you able to start and drive away or do you need to reset the maps to keep the engine running ?
The logic inside the ECU is designed in way that the CPU is not able to mess up the maps during cranking (or during any other voltage drops from the main battery).
What you maybe could do is: When you get home after a ride, don't turn the engine off and readout the maps. Do this maybe 3 times until the value in each cell doesn't change anymore, or at least not significantly. Wait until the problem appears, then compare this map with your previous ones. If they are nearly the same (especially around idle) we are looking at the wrong thing. If they are way apart the question is, what would force the ECU to correct these values to such an extent ?
Maybe that helps ?
BTW. After I put my original chip back in, the engine runs faultless since then.
Peter
Noblebenn said:
Sorry what does the 70ns mean and what’s the difference with the normal chip?
The normal chip is a DS1230Y-100+ which means it has "Read and write access times of 100ns" (nano seconds)The DS1230Y-70 has a 70ns read and write access time so can be used in faster applications.
I would imagine the timing on the ECU is designed to write on 100ns chips so using a faster one will work fine, but won't yield any extra benefits
HTH
crypto said:
It's hard to believe the chip is deteriorating. Inside the "chip" is only memory, a battery and some logic to switch between battery or external supply. In other words, the only thing that can deteriorate is the battery.
The 256k Nonvolatile SRAM is used to hold the 'adaptive' maps, and because it is battery backed up, it remembers them when the engine is turned offIt could be a possibility that the 'adaptive' maps are over time storing bad data which means when they are accessed the car ECU has adaptive mapping data that are contributing to the starting problems
Replacing the ECU / replacing the SRAM / flushing the engine Maps with a PC will just basically remove any existing 'adaptvie' maps, and then the engine starts with the non-adaptive Maps programmed into the ROM
My understanding is that the TVR use of the MBE ECU is different from others, in that it has these 'adaptive' maps that are stored over time to 'tune' the car - but ultimately these are based around the 'stored' maps in the ROM chip in the ECU
That's why burning a new ROM after the engine has been mapped will result in better results as you are starting from a tuned MAP and then adapting from that
Now.. as to what is causing the issues with your Maps I don't know - but you could replace the SRAM with a volatile RAM which means it is just cleared every time you turn the car off (RAM with no battery) - would be an interesting test to try and prove the above!
HTH
Edited by Juddder on Thursday 26th October 14:07
Juddder said:
The normal chip is a DS1230Y-100+ which means it has "Read and write access times of 100ns" (nano seconds)
The DS1230Y-70 has a 70ns read and write access time so can be used in faster applications.
I would imagine the timing on the ECU is designed to write on 100ns chips so using a faster one will work fine, but won't yield any extra benefits
HTH
The original chip in my Cerb's ECU was a DS1230AB-200, so I suspect any with an access time lower than 200ns will work fine.The DS1230Y-70 has a 70ns read and write access time so can be used in faster applications.
I would imagine the timing on the ECU is designed to write on 100ns chips so using a faster one will work fine, but won't yield any extra benefits
HTH
Thanks for the replies.
I am able to start the car again but it runs lumpy and it is difficult to drive until adaptive reset. However, on one or two occasions it has idled badly for several minutes and then sorted it self out.
Will try to read the maps as advised, is this done via laptop or the in car screen?
I am able to start the car again but it runs lumpy and it is difficult to drive until adaptive reset. However, on one or two occasions it has idled badly for several minutes and then sorted it self out.
Will try to read the maps as advised, is this done via laptop or the in car screen?
Gassing Station | Cerbera | Top of Page | What's New | My Stuff