Another MTX Fault - Jon Riglar's MTX

About original Memotech hardware.
jonriglar
Posts: 20
Joined: 14 Aug 2015 08:46

Re: Another MTX Fault - Jon Riglar's MTX

Post by jonriglar » 13 Oct 2015 21:43

Hello all,
Apologies for silence for a few weeks.
Have now 'scoped the memory chips.
I get input to all the memory chips (main and video) on Pins 2 (Data in)
I get output from all the main memory chips on pins 14 (data out)
I get NO output from all the video memory chips on pins 14 (data out)
Correspondingly I have no composite video signal on J13, Pin 5.

So I guess I'm back to a video memory issue, even though these chips have been changed.
Regards
Jon

Martin A
Posts: 406
Joined: 09 Nov 2013 21:03

Re: Another MTX Fault - Jon Riglar's MTX

Post by Martin A » 13 Oct 2015 23:53

I've never thought of scoping the VRAM pins, as interpreting the output is somewhat tricky.

So I've just had a go on the system being used to test/develop CFX, It's displaying the CFX's boot screen which is white text on dark blue

This was the trace for what the manual says is the RD2 line
Ads00010.jpg
Pin 14 of 7F slow trace
Ads00010.jpg (30.36 KiB) Viewed 3109 times
Ads00012.jpg
pin14 of 7F fast trace
Ads00012.jpg (31.89 KiB) Viewed 3109 times
The slow trace clearly show the differences between the active display part of the frame and the border. However the fast trace is showing each pulse as being high for 350 ns or so out of every 1500.

http://openmsx.org/vdp-vram-timing/vdp-timing-2.html Has more timing info about the VRAM, and says each read actually takes 372ns.

You didn't say whether your reading was stuck high or stuck low, But I'm wondering if a 1mhz scope could resolve that signal or whether it would appear as a low.

jonriglar
Posts: 20
Joined: 14 Aug 2015 08:46

Re: Another MTX Fault - Jon Riglar's MTX

Post by jonriglar » 14 Oct 2015 22:00

Hi Martin
Thanks for the info. I've got a 100MHz digital storage scope.
Pin 14 is stuck low on all the VRAM chips.
I've tried to isolate by removing one at a time, without any change.
The AD0-AD7 lines are buzzing up and down happily.
The CAS/RAS lines are also buzzing up and down.
It's just Pin 14 on each that's at ground.
I will double check voltages and for shorts.
Otherwise I'm wondering whether the replacement VRAMs are simply not compatible.

Martin A
Posts: 406
Joined: 09 Nov 2013 21:03

Re: Another MTX Fault - Jon Riglar's MTX

Post by Martin A » 14 Oct 2015 23:32

You can get away with using "modified" 64k DRAM chips if need be, they have the advantage of not needing -5v or 12 and also run cooler as a result

http://primrosebank.net/computers/mtx/p ... mavram.htm

I ran another experiment this evening, I faked a main memory fault by removing the OS ROM from the test bed MTX, and then scoped the VRAM on pin 14.

The wave form on the scope at 250ns was identical to the earlier test.

So, with the VDP not initialised and the black screen and tone coming from the TV, the VDP was reading the ram.

Interesting, and not what I thought I would see. I was expecting there to be no activity, as the display is blanked.

User avatar
Dave
Posts: 738
Joined: 11 Aug 2012 18:16
Contact:

Re: Another MTX Fault - Jon Riglar's MTX

Post by Dave » 14 Oct 2015 23:40

Hi Jon,

I can safely say that the VRAMs are definitely compatible, I have used them in a few MTXs with no problems.

If pin 14 (data out) on them all is low, it would appear that data is not being read.
You could look at the R/W pin (3), but I don't think it can be a video problem.

Even if the VRAMs were bad, the CPU would still execute the ROM code and you would get audible output if you pressed <CTRL><G>

regards
Dave

User avatar
Dave
Posts: 738
Joined: 11 Aug 2012 18:16
Contact:

Re: Another MTX Fault - Jon Riglar's MTX

Post by Dave » 15 Oct 2015 00:13

Jon,

can you post a high resolution picture of your computer board please?
Ideally, the solder side too, but, at least the component side

regards
Dave

User avatar
1024MAK
Posts: 555
Joined: 24 Dec 2012 03:01
Location: Looking forward to summer, in Somerset, UK

Re: Another MTX Fault - Jon Riglar's MTX

Post by 1024MAK » 15 Oct 2015 01:58

If the video DRAM is not being asked to output data, pin 14 will have no output data on it for all chips. So check pin 3 ( /W ).
Also if you 'scope is a four channel type, compare /RAS (4), /CAS (15), /W (3) and Q (14).

For the chip to output data, /RAS, /CAS need to be low and /W must be high.

Mark

jonriglar
Posts: 20
Joined: 14 Aug 2015 08:46

Re: Another MTX Fault - Jon Riglar's MTX

Post by jonriglar » 15 Oct 2015 21:28

Thanks for all the info :)
I'll get a photo done over the weekend.
A quick check of Pin 3 (/W) and it's high.
I've checked voltages/grounds and all ok.
/RAS (4), /CAS(15) and Q (14) merrily go up and down
Jon

jonriglar
Posts: 20
Joined: 14 Aug 2015 08:46

Re: Another MTX Fault - Jon Riglar's MTX

Post by jonriglar » 19 Oct 2015 22:03

Progress update!
There is some, thankfully.
I swapped out the VRAM for modified 4164 RAM and now I have output on Pin 14 of the chips.
I can also tune a full grey screen, which has a constant tone.
Tapping Control & G on the keyboard makes no difference.
At least it's semi-alive!!

Martin A
Posts: 406
Joined: 09 Nov 2013 21:03

Re: Another MTX Fault - Jon Riglar's MTX

Post by Martin A » 22 Oct 2015 20:15

Interesting result, the big difference between the 4116 and hacked 4164 is the voltages, the newer chip is 5v only.

It might be worth double checking the power transistor and regulators again.

Post Reply