EPROM flash and change

[Deviousdexter]

Hello

I hope this the correct place to post this, feel free to move it if not.

Please excuse my ignorance but i require some knowledge please. I have an old 8 bit computer which i believe needs the BIOS Rom replacing, my intention is to replace it with an Eprom and flash it myself. I have purchased a cheap programmer form China and downloaded some software which i believe to be compatible. However, nothing i look at seems to list the chip i am trying to flash.

The original chip is a Toshiba TMM23256P which i would like to replace.

The replacement is an AM27C256-120DC which i think is an AMD IC (??)
( This is UV Erasable ? Is it usable or do i need something else?)

Pics attached.

The programmer is a SP200S-V2.00 which connects via USB using a CH340 Driver. And i have ASprogrammer. NEOprogrammer and WLPro-V220 that seem to lauch ok but not detect or list the specific chip(s) i have.
Do i need a different chip, a different program(s) or is this possible with what i have? All i have in way of a UV light is a curing light for solder mask, is this useable or not ?

Any advise is greatfully received.

thanks

 

[betwixt]

Not sure exactly what model of computer you have but most MSX machines use Z80 processors so their clock speed is no faster than 6MHz and more likely 4MHZ. They never made faster Z80s, the Intel range overtook them in the computer market although they were and still are used in some industrial systems. That means the chance of it running with a slower EPROM is good. Can program one for you here if you like, no charge, it is highly unlikely I will live long enough to use my IC stocks, I've got thousands of varied devices that are collecting dust (anti-static dust of course!).

Brian.

 

[Deviousdexter]

Ok ,the plot thickens. I checked the few machines i still have and none have a parallel port. Easy, i'll grab one of Amazon !. It arrives, get fitted and setup. The Eprom software is expecting a LPT on 0x378, according to Device Manager it is using EFF8-EFFF or EFF0-EFF& but software says its on 0xEC00(??)

i have added this to an IO.ini as per https://github.com/IceMupppet/willem-eeprom-programmer/blob/master/README.md
This does not seem to work, i am also still getting errors from the program on launch to do with extracting io.sys (??? ,i mention it now as i had assumed this may be linked to the lack of LPT on my machine, or the fact it is 64 bit)
The board LEDs have been lit at various stages, VPP is currently on, as the board is connected to LPT.

Sorry about the size


The board also has a daughter board inthe connectors next to SP1, i have no idea what this does or if it should be there.
When launching the software, i get the following :-

The software then launches but cannot be closed without the "Privileged Instruction" warning and returning to the program ( i have to use task manager to quit the app).

I can select devices and change tabs without issue, but certain tasks cannot be executed showing the same behaviour as trying to quit.

I have tried a couple of compatibility modes to no avail.

Does anyone have any ideas ?

@betwixt - would like to explore a little further but would be very grateful if you have the time to do so, will gladly reimburse for any parts used.

thanks all

 

[betwixt]

Now you know why I haven't touched Windows in years!

The port address problem is caused by Windows virtualization of the ports. In the old days a port, whether printer, serial or even USB was addressed by its real electrical port number, unfortunately for us, when multi tasking programs came along they created a big problem. Two or more programs running simultaneously would be able to write to the port at the same time, or at least share it with data from each program becoming interwoven. For example, one program could be trying to print a document while another tried to print graphics and what came out of the port would be an indescribable mess. Windows gets around it by creating virtual ports, to each program they appear to be a real port but they are in fact a software emulation of the real thing. Each program gets its own instance of the port. A different program called a "spooler" checks to see if anything is pending printing and ensures each port is serviced in sequence but only when its predecessor has finished its job. That way documents are printed one after another. Sadly, your programmer is trying to address the physical hardware port at 0x278 or 0x378, not the virtual one that Windows has created. I'm sure there is a way to solve the problem but I have little experience with later versions of Windows, I gave up on it when XP was around and switched to Linux instead.

If you send me your name and address through http://www.atv-projects.com/Contact.php I will program an IC and send it to you FOC.

Brian.

 

[danadakk]

If you need a cheap UV source a UV power led might suffice....

https://www.aliexpress.us/w/wholesale-uv-led.html?spm=a2g0o.home.search.0

Be aware do not expose your eyes or skin to any of these high power LEDs,
exercise caution.

Light-Emitting Diodes (LEDS): Implications for Safety : Health Physics

more efficient than traditional lighting sources. LEDs emitting in the ultraviolet wavelength region have also become available and have made their way into consumer products. All these changes have led to a rise in concern for the safety of the optical radiation emissions from LEDs. Several in...

journals.lww.com

https://www.unmc.edu/ehs/safety/UVHazardsSafetyGuideline.pdf

If you need a cheap tool for spectral analysis : https://hackaday.com/2012/08/27/turning-a-webcam-into-a-spectrometer/
You would need a calibrated source to make measurements of absolute spectral power....


Regards, Dana.,

 

[Deviousdexter]

@betwixt - I have sent the info as requested. thanks again.

I hear what your saying about the address issue but some of the software in the GitHub is supposed to 'redirect' this address (from what i have read) and should allow this to function correctly. I never really put enough time or effort into Linux, which is a downfall on my part as i enjoyed playing around with RedHat many years ago when webpages were a buzz word. I have always tended to take what i need and research as far as needed to get it going. As always, i have my fingers in more pies than i have fingers. This project was meant to be a simple clean up and test of some old equipment,before selling it on to someone with more enthusiasm as i currently have, and hopefully get some enjoyment out of it. You are aware of the rabbit hole i went down from this thread.
Never the less, its all good for filling time and i undoubtedly will be in a similar predicament in due course

I will continue to 'play' and hopefully find a way to get things sorted or even find an older machine with a parallel port built in to repurpose. i've thrown so may away in the last 4 years to reclaim some space, it sooo annoying.

thanks go to anyone who has made a suggestion or linked info, it is really appreciated. Feel free to continue to offer advice as i will be looking into this as an ongoing project.

thanks

 

[betwixt]

Got the email thanks.

Most UV LEDs radiate at about 395nM despite what the adverts say. Undoubtedly they will erase an EPROM but so will white light given long enough.
Ideally you want a UV source near 250nM, much shorter wavelength than a normal UV LED. They are available but at sky high prices.
Mercury vapor tubes are still the best solution and give best value for money. Typical UV LED directly touching the EPROM window will take up to 24 hours, a Hg tube will take about 10 minutes and could erase maybe 10 at a time if lined up.

Whatever is used, DO NOT STARE at the UV source, almost all the emitted light is invisible and is also far more damaging to your eyes than the visible part of the spectrum.

Brian.