Post by thread:EPROM 16C73 ERASING TIME

Hi Group, I been using some EPROM "windowed" 16C73 chips. I am finding that over several cycles of erasing/programming (about 20) that the time to UV erase is becoming longer and longer. I have a few that will not erase at all. I have not been setting the code protect bit during programming. I have not ruled out the UV eraser but would think that it would impact all devices and not just a few. I also have not ruled out the PIC programmer either, the PICMATE, is the one I am using. Any suggestion? Paul

PDRUNEN@AOL.COM wrote: > I been using some EPROM "windowed" 16C73 chips. I am finding that over > several cycles of erasing/programming (about 20) that the time to UV erase is > becoming longer and longer. I have a few that will not erase at all. You too? Glad to see I'm not the only one with this problem. I have a UV eraser with a timer and I've had to go from 2 minutes for the first erases up to 20 now (after about 40 cycles). What is wrong with MicroChip's EPROM technology that it takes longer to erase as the cycle count rises? I see nothing in their spec sheet on EPROM life but this effect is absurd. The 16F87z series has a spec of 1000 cycles for the FLASH, so beware that you can't cycle it forever. > Any suggestion? Microchip doesn't know how to make EPROM? One could try baking the device at a high temperature (100C) for a hour or so to dissipate the trapped ions (worked for old 1702 EPROMs).

On Mon, 6 Mar 2000, Robert Rolf wrote: > spam_OUTPDRUNENTakeThisOuTAOL.COM wrote: > > I been using some EPROM "windowed" 16C73 chips. I am finding that over > > several cycles of erasing/programming (about 20) that the time to UV erase is > > becoming longer and longer. I have a few that will not erase at all. > You too? Glad to see I'm not the only one with this problem. > > I have a UV eraser with a timer and I've had to go from 2 minutes > for the first erases up to 20 now (after about 40 cycles). > > What is wrong with MicroChip's EPROM technology that it takes > longer to erase as the cycle count rises? I see nothing in their > spec sheet on EPROM life but this effect is absurd. > > The 16F87z series has a spec of 1000 cycles for the FLASH, so > beware that you can't cycle it forever. > > > Any suggestion? My eperience is refeer to standard eproms. There are 2 major rules to erase an eeprom: 1.insert chip in an aluminium folia any time you erase the chip and be soure that pins are made a good conexion with folia 2.use a powerfull lamp (100 to 500VA) and the smallest erasing time its possible( with testing at every 10...20 seconds of erasing time ) and keep the chip cool (its better if lamp have the desired radiation vawelenght for the quartz windows used on chip, and this is not the same at different chips) Eproms older than 5 years can be erased in 1...2 hours and can be written only after a ff writing was done. Vasile

Hi, excuse for the stupid remarks: you sure know the window must be ABSOLUTELY clean, if not then that problem can arise. Clean with alcohol, and fine cloth. I have also a 16C73, and I have absolutely no problem. I have a Conrad eraser, with a 3 Watt tube. However, I have also an eraser with a bulb and a filament in it. I can not guess how it works, but it does. Regards, Imre On Mon, 6 Mar 2000, Robert Rolf wrote: {Quote hidden}> .....PDRUNENKILLspam@spam@AOL.COM wrote: > > I been using some EPROM "windowed" 16C73 chips. I am finding that over > > several cycles of erasing/programming (about 20) that the time to UV erase is > > becoming longer and longer. I have a few that will not erase at all. > You too? Glad to see I'm not the only one with this problem. > > I have a UV eraser with a timer and I've had to go from 2 minutes > for the first erases up to 20 now (after about 40 cycles). > > What is wrong with MicroChip's EPROM technology that it takes > longer to erase as the cycle count rises? I see nothing in their > spec sheet on EPROM life but this effect is absurd. > > The 16F87z series has a spec of 1000 cycles for the FLASH, so > beware that you can't cycle it forever. > > > Any suggestion? > > Microchip doesn't know how to make EPROM? > > One could try baking the device at a high temperature (100C) > for a hour or so to dissipate the trapped ions (worked for > old 1702 EPROMs). > >

On Mon, 6 Mar 2000 00:20:09 -0700, you wrote: >PDRUNENKILLspamAOL.COM wrote: >> I been using some EPROM "windowed" 16C73 chips. I am finding that over >> several cycles of erasing/programming (about 20) that the time to UV erase is >> becoming longer and longer. I have a few that will not erase at all. >You too? Glad to see I'm not the only one with this problem. > >I have a UV eraser with a timer and I've had to go from 2 minutes >for the first erases up to 20 now (after about 40 cycles). Are you sure it's not your eraser tube ageing, or a dirty window? Apparently fingermarks block a lot of UV

At 10:46 PM 3/5/00 EST, you wrote: >Hi Group, > >I been using some EPROM "windowed" 16C73 chips. I am finding that over >several cycles of erasing/programming (about 20) that the time to UV erase is >becoming longer and longer. I have a few that will not erase at all. I have >not been setting the code protect bit during programming. > >I have not ruled out the UV eraser but would think that it would impact all >devices and not just a few. I also have not ruled out the PIC programmer >either, the PICMATE, is the one I am using. > >Any suggestion? >Paul > Paul, I've had similar experiences, and DID try cleaning UV windows and eraser tubes, new erasers, bug-free code, etc, etc, etc. I've more or less decided longer erasure times are a fact of life. [I also had a guy tell me, "if you weren't such a hacker, you wouldn't have to erase so much!" - I haven't talked to him recently]. Also, do you have a large value capacitor tied directly to the MCLR line? There was a discussion, related to this, last month on the piclist threads: Re: help me please!! .... AND Re: help me please!! (MCLR problems) Also, check out the comments in the "PIC Errata" section on my website (which started the controversy in those threads): http://www.sni.net/~oricom/piclinks.htm - Dan Michaels Oricom Technologies ===================