Assume you have a jump table in high memory. On powerup, the processor scans the table and jumps to the last non-FF address. Next time, you put the new code above the existing code in the address space and add a jump vector to the table. Or maybe overwirte earlier entries with 00 and jump to ghe first nonzero address.
So, is there anything in the chip that would prevent doing this? Would I have to rewrite the burner code to allow burning a non-blank part? Maybe this code already exists? thanks, mike
--
Return address is VALID.
500MHz Tek DSOscilloscope TDS540 $2200
http://nm7u.tripod.com/homepage/te.html
Wanted, 12.1" LCD for Gateway Solo 5300. Samsung LT121SU-121
Bunch of stuff For Sale and Wanted at the link below.
http://www.geocities.com/SiliconValley/Monitor/4710/
And I'm too lazy to go look it up, but isn't 0x00 a NOP? I remember reading a trick to reprogram the OTP chips by leaving a gap at the beginning (of 0xFF) and starting with a goto your code. Then each time you reprogram you turn the old goto into a nop and put a new one after it. I could be backwards and nop is 0xFF, but then you just put the first goto at the end of the gap and work backwards instead.
I'm also too lazy to go price them but I bet a 12F629 is cheaper than a 12C509 now.
No problem to do this, every decent programmer will jut ask if you want to proceed and burn non-blank parts. By the way Microchip OTP and Flash uC prices are almost the same so this would be applicable if you have a lot of these old OTP chips.
12C508 have only 512 words of program memory, so I doubt is it worth the troubles to re-compile your code at different memory offsets etc, istead to buy one Flash part and re-program it thousands of times.
Best regards Tsvetan
--
PCB prototypes for $26 at http://run.to/pcb(http://www.olimex.com/pcb)
PCB any volume assembly (http://www.olimex.com/pcb/protoa.html)
Development boards for ARM, AVR, PIC, and MSP430
(http://www.olimex.com/dev)
Most people do the first N units with an EEPROM parts and then go to OTP when they are sure that it works. Do you have a special application that makes you expect changes? Or do you have a problem with your process that results in poor requirements and inadequate testing?
On that chip, NOP is all zeros, check the data sheet, but I believe that it starts out all 0s and you can write only 1s. [
formatting link
]
It comes out of reset by executing whatever is a address zero with page zero selected, so you can easily put in a bunch of NOPs followed by a jump to your code, and later put in another jump that gets executed first.
Your burner may choke on programming a non-blank part. Check with the manufacturer.
Also, you can erase OTP parts by using X-Rays instead of UV. Make friends with a dentist.
There are even 12F508 parts available now (90 cents in 25's).
Best regards, Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
About 10 years ago I've programmed OTP micro's for a prototype and shipped them from the Netherlands to Sweden. Customer complained - they failed to do anything. Programmed a few more, tested them: OK. Sent them off. Same complaint. Customer returns chips - they were empty and fully reprogrammable. We sent another programmed and tested lot, now wrapped in Alu foil and with a big sticker "DO NOT XRAY - SENSITIVE DEVICES". All devices worked on arrival...
BTW: The dentist's Xray machine tube wears out relatively fast. I'm not sure what the real cost of erasing would be...
That appears to be evidence that x-rays can erase OTP parts - something that we already know to be true. It sheds no light upon the claim that x-rays *damage* OTP parts. I am still looking for evidence to support that particular claim.
Also, there have been many laptops with EPROMs in them going through many airport x-ray machines for many years - some of them many times - and the laptops still boot. I would suspect that your programmer was doing a marginal job.
Many years ago I had a big pile of OTP 27C512. That was a leftover from the upgrades of the EPSON printers. The unused memory in those parts could be programmed just fine.
I tried to reuse those 27C512 by erasing them with the X rays. No luck at all. The parts were partially damaged before getting completely erased. I spent several days trying to find the optimal voltage, current and exposure settings.
Indeed, the conventional 2764 with UV window was erased with X rays with no problems at all.
Perhaps the silicon in the OTP parts is different from the ordinary parts. It is not intended for multiple reprogramming.
The difference between OTP and UV eraseable is just the package the die is the SAME! Without the quartz window UV will not penetrate the ceramic or plastic casing. A plain casing is cheaper to produce than one with a quartz window.
One product many years ago was always giving problems on site, which was tracked doen to the boards that failed had a quartz window. As the customer had Beta radiation sources about that even penetrated the PC case these cards would scramble themselves and generally had to be unplugged and left for two days before being reused. We swapped out the eraseable parts for OTP and the customer had no further problems.
--
Paul Carpenter | paul@pcserviceselectronics.co.uk
PC Services
GNU H8 & mailing list info
For those web sites you hate
That is a guesswork. How do you know? It seems logical to use a lower quality silicon for the parts not intended for reprogramming, isn't it? My practical experience shows that it is not a good idea to erase OTPs with X-ray machine.
Over the years, various discussions with distributors and manufacturers, and in some cases past datasheets.
You can test for speed of access, how are you going to test on wafer how many times it is likely to programme, in a QUICK way. Many manufacturers even on Flash parts have been conservative on how many times their devices can be reprogrammed to cope with worst case situations. Many have changed their specs from 100 times to 1000s of times from experience and more extensive testing over long periods of time.
In reality it is easier to test the wafer that on reprogramming one or more times, the cells respond and get to correct levels on wafer/die test points for ALL types, then choose how to package as they do for all other devices.
It is more logical to manufacture and test for one die configuration and spec, and it is more cost effective. Even RAM parts that notoriously had many speed variants in the past to overcome yield problems and low margins are being produced with less speed variants, to cope with the other logistics problems of storing and handling speed variants, which adds to the costs.
Plastic/ceramic OTP packaging is cheaper as more parts are packaged that way. Windowed products are special cases and hence a different packaging process(es).
That maybe so which is different to the point I answered. Ever considered how X-Ray transparent ceramic/plastic is compared to quartz, let alone any heat dissipation or other characteristics that may be different.
Anyway why bother X-Ray work for the MAJORITY of people is very costly and expensive, let alone hazardous.
--
Paul Carpenter | paul@pcserviceselectronics.co.uk
PC Services
GNU H8 & mailing list info
For those web sites you hate
We had an incident years back where a series of OTP processors failed in the field. The manufacturer denied that this was possible. We provided samples for them to test. Ultimately, they confirmed our claims and replaced the processors. One of the things that came out of this was a couple of OTP parts that were converted to UV Erasable parts. They used fuming Nitric acid dripped onto the part to dissolve the PLCC encapsulation over the die. This formed a crater around the die, but left everything else intact. These parts could then be erased by UV. We added a quartz window and Indium bond to one and used it as a UV erasable part for verification. Still have it around here.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.