Discussion
The work of going through the ecus continues.
Today I’ve extracted the pcb from the mbe941. One of the screws for the lid was missing and water has been getting in.
Amazingly the unit still boots and diagnostics comes out.
However, as can be seen water has attacked the sockets for the nvram and eprom. Given this is for a car anybody have recommendations for replacement sockets?
These turned ones certainly seem upto the job
Today I’ve extracted the pcb from the mbe941. One of the screws for the lid was missing and water has been getting in.
Amazingly the unit still boots and diagnostics comes out.
However, as can be seen water has attacked the sockets for the nvram and eprom. Given this is for a car anybody have recommendations for replacement sockets?
These turned ones certainly seem upto the job
I’ve now bought a universal programmer and after arguing with Windows 11, it’s working.
I can read the eprom. Is there a list of known EPROMs and their checksums?
I’ve tried reading the nvram and it just gives 0xAA for every location. This doesn’t seem right. The pins are quite corroded, so I’ll give them another clean.
I can read the eprom. Is there a list of known EPROMs and their checksums?
I’ve tried reading the nvram and it just gives 0xAA for every location. This doesn’t seem right. The pins are quite corroded, so I’ll give them another clean.
The nvram has a lithium power source (I assume they mean battery) inside. The nvram will retain data as long as it’s powered. If the power is removed, then the lithium power source takes over. The data sheet says the power source is good for a minimum of 10 years once it’s activated.
I’ve now invested further and reprogrammed the chip. I’ll try and read it again and see if it still remembers.
Given the water damage, I’m going to get a replacement along with a fresh eprom.
I’ve now invested further and reprogrammed the chip. I’ll try and read it again and see if it still remembers.
Given the water damage, I’m going to get a replacement along with a fresh eprom.
Yes, I understood that, (my understanding is that the Nvram burns a link on it's first power up) but after that, things like a flat 12v battery, simply by turning on the ignition the car will see a usable 3v and takes the 3v. I then really would not expect it to hold any adaptive's. I would go as much as saying that it is indicative of good/poor servicing (our fully documented Tamora had it's first replacement at 18 years, 21k and having had 4 12v batteries, every journey/stop,restart, started with a reset). I have a bluetooth to my phone to warn me (all my cars, not just the TVR's) of a battery low and then recharge prior to even turning on the ignition A@
Edited by Adrian@ on Wednesday 15th October 22:35
From my reading of the various wiring diagrams, I think the ECU gets a permanent feed from the car battery which I'm assuming the MBE then supplies power to the NVRAM.
From reading the datasheet, the lithium power source is dormant when they (the nvram chips) come out of the factory. The first time external power is applied then the power source is enabled. If there is external power applied the internal lithium power is not used.
The datasheet says the lithium power is good for a min of 10 years of use - I read that as accumulated use when external power is not applied. i.e if the car has a good battery, which is kept connected and charged then the NVRAM chip will retain settings for a very long time.
As far as I'm aware, the Cerbera this chip came from has sat unloved for 13 years or so.
As a side note, it's interesting to see that the NVRAM chips are used in a lot of electronics related test gear from the 80's and 90's. As they have started to fail, people are building replacements based around modern FRAM chips. I wonder if anyone has tried one in a TVR?
From reading the datasheet, the lithium power source is dormant when they (the nvram chips) come out of the factory. The first time external power is applied then the power source is enabled. If there is external power applied the internal lithium power is not used.
The datasheet says the lithium power is good for a min of 10 years of use - I read that as accumulated use when external power is not applied. i.e if the car has a good battery, which is kept connected and charged then the NVRAM chip will retain settings for a very long time.
As far as I'm aware, the Cerbera this chip came from has sat unloved for 13 years or so.
As a side note, it's interesting to see that the NVRAM chips are used in a lot of electronics related test gear from the 80's and 90's. As they have started to fail, people are building replacements based around modern FRAM chips. I wonder if anyone has tried one in a TVR?
That about sums it up, (better than I could of, because I have a range of TVR's 71-2003 and only two with MBE) and I have a best mate calibration expert (Hydrogen engines these days) that I call on to help in these matters. Now I have to look up FRAM! Nice work/other drawings BTW. A@
Now that would interesting ...a piggy back board, that would take the battery out of the equation.
You must of realised by now that MBE have zero want or desire to get involved in the anyway. Even my mate, who was involved in taking Noble to serious BHP got blanked when I asked him to talk to them about TVR's NVRAM when I first got the Tamora, at the time, I didn't want to delve deep than a replacement (which cured the issues I was having), as I was taking a 'curve' on a T car (I am 4 years along that curve, having this week replaced with new/refurbished ALL the steering components). If it the way you are going, I would be interested. A@
Edited to add...perhaps a replaceable battery. https://www.youtube.com/watch?v=xBvw1TLHyqM
You must of realised by now that MBE have zero want or desire to get involved in the anyway. Even my mate, who was involved in taking Noble to serious BHP got blanked when I asked him to talk to them about TVR's NVRAM when I first got the Tamora, at the time, I didn't want to delve deep than a replacement (which cured the issues I was having), as I was taking a 'curve' on a T car (I am 4 years along that curve, having this week replaced with new/refurbished ALL the steering components). If it the way you are going, I would be interested. A@
Edited to add...perhaps a replaceable battery. https://www.youtube.com/watch?v=xBvw1TLHyqM
Edited by Adrian@ on Saturday 18th October 07:12
I’ve yet it investigate further, but there are projects that put a suitable FRAM on a pcb with pin headers which has the same layout as the Dallas chips.
It would seem the FRAMs have very similar layout and electrical requirements as the original chips because the pcbs only have one extra component on them.
It would seem the FRAMs have very similar layout and electrical requirements as the original chips because the pcbs only have one extra component on them.
Markb139 said:
The data sheet says the power source is good for a minimum of 10 years once it s activated.
I store the car with the battery disconnected. From my experience a new Dallas chip loses all data within 6 months.BUT - this is not a problem. Remember that the Dallas chip stores constantly changing data - depending on the style of your driving. The data is adaptive and designed to change. Also, if you tune the car every 6 months to a year - you should be zeroing this data in any event.
Let's assume you've just tuned the car and the adaptives are all zero. Go for a drive and within half an hour they will be populating again. Same applies for a chip that has leaked it's data away.
While replacing the Dallas chip with an FRAM might seem like a good idea to maintain the data, I'm not sure that in reality you'll be gaining much from the swap.
notaping said:
I store the car with the battery disconnected. From my experience a new Dallas chip loses all data within 6 months.
BUT - this is not a problem. Remember that the Dallas chip stores constantly changing data - depending on the style of your driving. The data is adaptive and designed to change. Also, if you tune the car every 6 months to a year - you should be zeroing this data in any event.
Let's assume you've just tuned the car and the adaptives are all zero. Go for a drive and within half an hour they will be populating again. Same applies for a chip that has leaked it's data away.
While replacing the Dallas chip with an FRAM might seem like a good idea to maintain the data, I'm not sure that in reality you'll be gaining much from the swap.
BUT - this is not a problem. Remember that the Dallas chip stores constantly changing data - depending on the style of your driving. The data is adaptive and designed to change. Also, if you tune the car every 6 months to a year - you should be zeroing this data in any event.
Let's assume you've just tuned the car and the adaptives are all zero. Go for a drive and within half an hour they will be populating again. Same applies for a chip that has leaked it's data away.
While replacing the Dallas chip with an FRAM might seem like a good idea to maintain the data, I'm not sure that in reality you'll be gaining much from the swap.
What is your thoughts on the watch battery mod. (a couple of quid rather than 40.00)
I could not 'sell' this to Joolz (it is my wifes car) that the adaptives were going to come back in 30 minutes, when, if she stopped the car at 29 minutes the 30 minutes started again (it was 18 an years old Nvram at the time).
(she currently only has to plug in and to reset the ECU when she calls in to unknown petrol stations on long trips).
A@
Edited by Adrian@ on Saturday 18th October 09:58
Markb139 said:
As a side note, it's interesting to see that the NVRAM chips are used in a lot of electronics related test gear from the 80's and 90's. As they have started to fail, people are building replacements based around modern FRAM chips. I wonder if anyone has tried one in a TVR?
Now I look is this what you are talking about (I do like the battery mod though, as I have my old chip and the battery holders!)...A@https://www.pcbway.com/project/shareproject/Dallas...
Edited by Adrian@ on Saturday 18th October 11:54
The battery mod looks fiddly. While it might work, I'd not be willing to risk a chip - even an old one. Also, for me - I'm not concerned about retaining the adaptive values. My engine actually runs better off the base map when the adaptives are zero and I frequently zero them when out driving, especially if I'm stuck in an endless chain of roadworks etc. I find the idle gets a lot smoother.
As for the FRAM / FeRAM - it appears that the read process is destructive necessitating the need for write after read architecture. I can't see them being at all compatible with the MBE firmware.
As for the FRAM / FeRAM - it appears that the read process is destructive necessitating the need for write after read architecture. I can't see them being at all compatible with the MBE firmware.
Thanks for that valuable info, you are running the V8 is a thought (I have not owned a TVR V8, our 6 runs poorly until the car settles after a reset).
I might play with my old chip (I popped a meter on it and it is dead, BUT I knew that years ago), it has been sat here for the last 4 years AND I have 6 years before the car sort of needs a change (I am doubt I will get to the point of a flat 12v battery that might stress the chip battery).
A@
I might play with my old chip (I popped a meter on it and it is dead, BUT I knew that years ago), it has been sat here for the last 4 years AND I have 6 years before the car sort of needs a change (I am doubt I will get to the point of a flat 12v battery that might stress the chip battery).
A@
Inbox said:
That pcba does not have a conformal coating on it, the manufacturer was relying on the enclosure to protect it so nothing to reinstate.
Again valuable information, one of the 'failures' (it stopped the remote from working on the alarm, after several incidents where the house cameras 'saw' the indicators flashing randomly) was verdigris on a soldered joint on the TVR mod to bring the alarm push/override across the car (two wires joined together). At that time, I simply cleaned and re-taped over the joint. I was looking for a second wire (I already have one) to re install the wire across the car.
NOW, I have to buy in some conformal coating to seal the PCB in that joint. A@
Edited by Adrian@ on Monday 20th October 08:23
I bought a Ramtron FRAM part number FM1808-70-PG to try in place of the Dallas IC. However, before I tried it in my ECU I wanted to test program it on my TL866A programmer and ideally program it with my current Dallas IC data. I can read and write Dallas ICs but I couldn't get my programmer to program the FM1808.
The TL866A I have doesn't natively support the FM1808 and I thought I could use the parameters to read/write it by using the DS1230AB settings. However, I can read but not write. I suspect the FM1808 needs 5V to write and the Dallas 3.3V but I don't have the datasheet to hand to confirm.
Anyway, I've parked this idea for now until I can read/write the FRAM before installing in my MBE ECU. I don't want to risk my ECU, although a dead IC shouldn't damage it. The FM1808 was <£10 from China.
The TL866A I have doesn't natively support the FM1808 and I thought I could use the parameters to read/write it by using the DS1230AB settings. However, I can read but not write. I suspect the FM1808 needs 5V to write and the Dallas 3.3V but I don't have the datasheet to hand to confirm.
Anyway, I've parked this idea for now until I can read/write the FRAM before installing in my MBE ECU. I don't want to risk my ECU, although a dead IC shouldn't damage it. The FM1808 was <£10 from China.
Forum | Cerbera | Top of Page | What's New | My Stuff