The Dos and Don'ts of using an EPROM as PLA in a C64.

Disk drives, Monitors, SuperCPU etc.
banman
Member
Member
Posts: 48
Joined: Sat Jun 15, 2019 12:21 am
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by banman » Wed Oct 09, 2019 11:18 am

Hi eslapion,


Thank you for this information... :D

Here is an observation I made with some changes to the oscilloscope settings.

I hope I am getting closer to the correct result..... :-)



Apologies for the mistake in the video.. It should be graticule scale . :oops:





User avatar
eslapion
Member
Member
Posts: 819
Joined: Mon Jul 20, 2015 10:11 am
Location: Canada
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by eslapion » Fri Oct 11, 2019 9:24 am

1. Your board is assy 250407, the other number you've read is the schematics.

2. If you read 500mV while the probe is getting 5V that's because you set the probe to 10X but didn't tell the scope.

3. There is no such thing as 'negative' slope. There is rising or falling slope trigger.

Otherwise, you're getting closer.

Good read: https://download.tek.com/manual/070633902web.pdf

https://www.tek.com/manual/2430a-service-manual
Wealth, like happiness, is never attained directly. It comes as a by-product of providing a useful service. -Harland D. Sanders

banman
Member
Member
Posts: 48
Joined: Sat Jun 15, 2019 12:21 am
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by banman » Sat Oct 12, 2019 2:41 am

Hi eslpion,
eslapion wrote:
Fri Oct 11, 2019 9:24 am
Your board is assy 250407, the other number you've read is the schematics.

Oops, apologies for that silly error..... :oops: :oops: :oops:



You are correct well spotted. It is indeed a C64 assy 250407 board. The Winbond chip behaves very badly in this configuration.

I appreciate the corrections on the terminology used regarding the oscilloscope. :D


Thank you very much for that information on the Tektronix 2430A oscilloscope.

I have been wanting something like this... The manual I have is a little beyond my grasp at this stage.... :D

This will take me some time to digest it all ..... I will start off with the Triggering section, that's what I need to get right....



I believe the Winbond W27C512 EEPROM will trigger substantially lower than the 3 v Trigger I set on the oscilloscope in the video. I will check it out properly and get back to you....


I am working on getting the voltage setting into the normal volts range and not the mV range....


I think my next step is to get ROMH! and ROML! to display on the screen....

User avatar
eslapion
Member
Member
Posts: 819
Joined: Mon Jul 20, 2015 10:11 am
Location: Canada
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by eslapion » Sat Oct 12, 2019 7:09 am

banman wrote:
Sat Oct 12, 2019 2:41 am
I think my next step is to get ROMH! and ROML! to display on the screen....
Assuming you have a problematic (E)PROM used as PLA which generates glitches, you have to have some experience to really exhibit them well as I have in the early posts of this thread. I suggest you experience with your scope with regular signals to start with.

In the cartridge port, pin 6 is the dot clock which cycles at about 8 MHz (a bit less on PAL system, a bit more on NTSC systems).

Pin E on the lower row of pins is the system clock at about 1MHz (exactly 1/8 the dot clock). You can also get this signal on pin 1 of the 6510 CPU.

On the VIC-II, you can get the dot clock (same as cart port) on pin 22 and the color clock on pin 21 (17.7344MHz for PAL and 14.31818MHz for NTSC)
See: https://ist.uwaterloo.ca/~schepers/MJK/vic2.html
Wealth, like happiness, is never attained directly. It comes as a by-product of providing a useful service. -Harland D. Sanders

banman
Member
Member
Posts: 48
Joined: Sat Jun 15, 2019 12:21 am
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by banman » Mon Oct 14, 2019 5:33 am

Hi eslapion,




Am I correct in thinking the 10x setting is used to reduce the loading on the circuit under test? This would allow the circuit under test to perform with out being interfered with by the oscilloscope.




I think I have found out the reason why the oscilloscope won't auto range when I enable 10X setting on the probe.

From what I have researched the issue is with some generic probes. Such as one I am using. The old factory ones were different (more electronics in them i think). I noticed an electrically separate ring/collar at the base of the BNC connector. I think this collar comes into contact with some part of the factory probe signalling the oscilloscope into the correct range.


I was doing some research on my issue on the internet.

I found this article



https://electronics.stackexchange.com/q ... ded-probes




The options I have at the moment are:

1. find a physical work around on the generic probes I have

2. find replacement probes that are factory equivalents

or

3. do the conversion in my head (upscale the values I see on the screen by a factor of 10).

For the time being option 3 is my only real avenue. I just have to remember that all the readings I get in 10X mode need to increased by a factor of 10.





I am finding that the C64 Assy 250425 board is more fault tolerant than some of the older boards I have.


I might try using the C64 assy 250407 board for obtaining readings. I observed good clear readings using the Winbond W27C512 EEPROM as the PLA substitute in this 250407 board.

eslapion wrote:
Sat Oct 12, 2019 7:09 am
I suggest you experience with your scope with regular signals to start with.

Very good idea! :)

I'll get on to it....

User avatar
eslapion
Member
Member
Posts: 819
Joined: Mon Jul 20, 2015 10:11 am
Location: Canada
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by eslapion » Tue Oct 15, 2019 11:00 am

banman wrote:
Mon Oct 14, 2019 5:33 am
Am I correct in thinking the 10x setting is used to reduce the loading on the circuit under test? This would allow the circuit under test to perform with out being interfered with by the oscilloscope.
It's only one of multiple reasons.

See here: https://www.electronics-notes.com/artic ... probes.php

From what I have researched the issue is with some generic probes. Such as one I am using. The old factory ones were different (more electronics in them i think). I noticed an electrically separate ring/collar at the base of the BNC connector. I think this collar comes into contact with some part of the factory probe signalling the oscilloscope into the correct range.
Yup! This ring tells what type of probe you have.

My TDS3014B has the detector for this ring but the channel settings always allows you to manually tell what type of probe you have if the ring isn't on your specific probe.
I have a 10X and 100X probes with the automatic setting collar and answer 3 seems to be correct.

The options I have at the moment are:

1. find a physical work around on the generic probes I have

2. find replacement probes that are factory equivalents

or

3. do the conversion in my head (upscale the values I see on the screen by a factor of 10).

For the time being option 3 is my only real avenue. I just have to remember that all the readings I get in 10X mode need to increased by a factor of 10.
AFAIK, all Tektronix scopes will allow you to set the probe type manually in the channel settings.

BTW, the problem you're having with your probes right now is one of the reasons why I told you to stay away from old scopes of this generation. Presently, there is a way to tell the scope you have 10X probes but you don't know where this setting is and I can't search through its functions because it's not in front of me. For all TDSxxx scopes, you have the MENU button below 'VERTICAL' panel.
I am finding that the C64 Assy 250425 board is more fault tolerant than some of the older boards I have.

I might try using the C64 assy 250407 board for obtaining readings. I observed good clear readings using the Winbond W27C512 EEPROM as the PLA substitute in this 250407 board.
Older boards 250407 are more sensitive to problems with the PLA by virtue of having slower chips (less timing margin left in each cycle) and by sometimes not having the R42-C204 RC filter on CASRAM.

Added edit:
Another option is to get other probes which have the coding built-in. It will always be good if you get another scope in the future.
This specific one is rather inexpensive: https://uk.rs-online.com/web/p/oscillos ... s/1799561/
Wealth, like happiness, is never attained directly. It comes as a by-product of providing a useful service. -Harland D. Sanders

banman
Member
Member
Posts: 48
Joined: Sat Jun 15, 2019 12:21 am
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by banman » Wed Oct 16, 2019 8:48 am

Hi eslapion,


Thank you very much for all that great information. Much appreciated! :D

You have actually given me a great idea regarding the probe work around.... ;) Let me think it thru and I'll get back to you...

I have been testing a few points on the C64 board as you suggest...



I am getting some good observations....



Here is a test on pin 1 of the 6510 CPU on a C64....





I am not too sure if the numbers I am getting are because the test equipment is out or the C64 is out.. All the C64''s I am testing are PAL versions....

banman
Member
Member
Posts: 48
Joined: Sat Jun 15, 2019 12:21 am
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by banman » Wed Oct 16, 2019 11:35 am

Hi eslapion,


Here is one solution I came to regarding getting the 10X probe to display the correct voltage.




I have found that the wave forms are basically the same just the voltages are altered to correctly reflect what is under test.

User avatar
eslapion
Member
Member
Posts: 819
Joined: Mon Jul 20, 2015 10:11 am
Location: Canada
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by eslapion » Wed Oct 16, 2019 8:17 pm

banman wrote:
Wed Oct 16, 2019 8:48 am
I am not too sure if the numbers I am getting are because the test equipment is out or the C64 is out.. All the C64''s I am testing are PAL versions....
Looks perfectly ok to me. You've tapped into the CPU clock which is a nice square wave.

The minimum, accounting for the 10X probe not recognized by the scope is 40mV and the maximum is 3.76V (displayed as 376mV). The PAL CPU clock is a tad lower than 1MHz and the number you have seems correct.

What some ice cream and cherry to top this ?

Here is one solution I came to regarding getting the 10X probe to display the correct voltage.

I have found that the wave forms are basically the same just the voltages are altered to correctly reflect what is under test.
In the video you say you used a 10 Ohms resistor. I sure hope that's not the case as it would tell the scope you have a 1000X probe. 10Kohms (10 000 ohms) is within the range for a 10X probe.

Added edit: You later say in the video it's a 10kOhms resistor so... nice fix.
Remember you don't have to get the ground from the probe itself. You can take it from any other BNC connector so you could use your plastic ring to install a switch between the external sensing ring to the ground to switch back and forth between 1X and 10X. Personally, I never use 1X because the bandwidth is too low in that mode.
Wealth, like happiness, is never attained directly. It comes as a by-product of providing a useful service. -Harland D. Sanders

banman
Member
Member
Posts: 48
Joined: Sat Jun 15, 2019 12:21 am
Contact:

Re: The Dos and Don'ts of using an EPROM as PLA in a C64.

Post by banman » Thu Oct 17, 2019 4:42 am

Hi eslapion,


Oops..... :oops: :oops:

My Bad regarding the resistor rating.... I should have said 10M Ohm resistor not a 10 ohm resistor. Apologizes for that error...

eslapion wrote:
Wed Oct 16, 2019 8:17 pm

Looks perfectly ok to me. You've tapped into the CPU clock which is a nice square wave.

Oh, that is good I got the readings right.

eslapion wrote:
Wed Oct 16, 2019 8:17 pm
Remember you don't have to get the ground from the probe itself. You can take it from any other BNC connector so you could use your plastic ring to install a switch between the external sensing ring to the ground to switch back and forth between 1X and 10X. Personally, I never use 1X because the bandwidth is too low in that mode.

Great idea! I didn't think of that.

I will do another run of the Dodgy Winbond W27C512 EEPROM PLA substitute with the correct voltage settings.

Post Reply Previous topicNext topic

Who is online

Users browsing this forum: No registered users and 1 guest