A problematic Oric 1

If you want to ask questions about how the machine works, peculiar details, the differences between models, here it is !
How to program the oric hardware (VIA, FDC, ...) is also welcome.
rh74
2nd Star Corporal
Posts: 20
Joined: Mon May 17, 2021 8:53 am

Re: A problematic Oric 1

Post by rh74 »

Chema wrote:
Thu May 27, 2021 9:49 pm
You may want to check the main signals in the CPU. A stuck IRQ or RESET, bad clock in or clock out... Just in case...
Yes I've checked that, but now I am a little more aware about what to check exactly... Noel's C64 provided some hints and I'm going through some 6502 literature to have a clear idea about what should be a correct signal. I like challenges... :D

rh74
2nd Star Corporal
Posts: 20
Joined: Mon May 17, 2021 8:53 am

Re: A problematic Oric 1

Post by rh74 »

Hey there! Here some news to talk about. Finally I had some time to sit and do some oscilloscope measures. I found the PIN 39 on ULA is emitting garbage. I measured it a few times but the result is always the same: just noise instead of a square wave. I've tested the Atmos for comparison, signals are completely different. Now the hard part. I've exchanged the 10017 with the Atmos one, with the other Oric-1 and with a spare 10017 I have. Same defect.

So now... A couple of questions for you experts:
  • Pin 39 is intended for pin 14 on IC20 (74257). I didn't tried to exchange IC20 and IC8, I will do it asap to check if the problem persists. Anyway, could the target IC be the source of problems?
  • Here on this obscure site ( :D ): https://wiki.defence-force.org/doku.php ... :ula_10017 I've found a reversed engineering scheme of 10017, it states that the MA(x) are mirrors of A08 and above. Is there a map or can I just consider MA7 as a mirror of A15? What does it mean a mirror? I've measured them on the Atmos and It seems the signal is slightly different and amplified. Is there somewhere a more detailed 10017 datasheet?
  • I've checked the signal from CPU to the ULA and they seems ok Atmos/Oric-1. It seems something is not working after PIN 39, but no other components are involved and the IC20 should be ok if I trust two different tests... Any hints?
  • Any other suggestion?
Guys if I can manage to solve this, consider a payed beer whenever you'll step Italy.

rh74
2nd Star Corporal
Posts: 20
Joined: Mon May 17, 2021 8:53 am

Re: A problematic Oric 1

Post by rh74 »

Ah forgot to mention... The noising signal from PIN 39 is combined to the output of PIN 16 (6502) in pin 13 and 14 of IC20. The combined signal is then addressed to PIN 9 (A7) of RAM. Is where I initially found the missing square wave. Now afaik both are combined in two logical AND port and then in a logical OR. Since other PIN on RAM sourcing from the IC20 are fine, I think everything but PIN 39 - PIN 14 is failing. Now it is just matter of finding why... In the meantime I will search the basement for a spare 74LS257. I guess I can borrow it from the Apple ][, I'm pretty sure it has it.

User avatar
iss
Wing Commander
Posts: 1277
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: A problematic Oric 1

Post by iss »

Ciao, @rh47, the beer is very good way for motivation!
Although in Italy are many many other delicious things to try and left the beer as table-water replacement :D

Now to the Oric: measuring the signal at ULA's pin 39 (i.e. pin 14 of IC20) with oscilloscope is bit tricky - I found that if the probe is x1 then starts the mess the CPU stops and pin 14 of IC20 goes permanently high. But switching the probe to x10 (i.e. higher impedance) you can get good results.

Here is how I've connected the probes (yellow clip to pin39 of ULA (i.e. pin13 of IC20) and blue to pin 14 of IC20):
20210603_195427.jpg
... and what I captured - normal signals, don't expect to be very rectangular :)
(yellow trace = yellow clip, blue trace = blue clip, of course)
20210603_195506.png
Else, changing IC8, IC20 and IC21 is not bad idea.
One other thing - you can carefully cleanup on PCB's solder side the ULAs pins from old resin remains with alcohol - there can be shorts under it, if this is not the problem then at-least the PCB will look nicer :).

rh74
2nd Star Corporal
Posts: 20
Joined: Mon May 17, 2021 8:53 am

Re: A problematic Oric 1

Post by rh74 »

iss wrote:
Thu Jun 03, 2021 7:37 pm
...and left the beer as table-water replacement :D
Good idea...

So here a quick update about the quest for Oric 1.

I removed both IC20 and IC8 and measured PIN 39 of ULA. The signal is there. As soon as I inserted the IC20 back the signal gone
IMG_20210606_110757.jpg
I'm still wondering how I didn't noticed before, but here you are the super bad soldering action:
IMG_20210606_111058.jpg
In circle the PIN 14 of IC 20. I removed the superfluous tin and I was very happy to finally found the problem. Tested again with the oscilloscope, now the signal is running on PIN 39... But not on IC 20. :(

Swapped IC20 and IC8, the signal is now everywhere... Still the result is not changed. I have the checkered screen of RAM fault. I guess I will remove socks and replace them with something better and I will go through all the soldering. BTW do you know any technique to restore track isolation? In the close up you can see the green cover is missing on some tracks part. I'm wondering if this could be the problem. Another super bad solder could hide in the PCB.

User avatar
iss
Wing Commander
Posts: 1277
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: A problematic Oric 1

Post by iss »

rh74 wrote:
Mon Jun 07, 2021 11:02 am
BTW do you know any technique to restore track isolation?
On youtube are some videos about solder mask restoration (like this random one). I personally never tried such UV coating. Another easy option can be GF's nail polish :).

User avatar
Chema
Game master
Posts: 2809
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: A problematic Oric 1

Post by Chema »

Man, this one is tricky indeed! But you are closer... I have little doubt this Oric will be revived in the end!

rh74
2nd Star Corporal
Posts: 20
Joined: Mon May 17, 2021 8:53 am

Re: A problematic Oric 1

Post by rh74 »

iss wrote:
Mon Jun 07, 2021 12:41 pm
Another easy option can be GF's nail polish :).
Or wife's... Once I've seen a marker style technique. You were able to cover tracks just following them with a marker. An easy way indeed.
Chema wrote:
Mon Jun 07, 2021 8:38 pm
I have little doubt this Oric will be revived in the end!
I'm not yet defeated... I would be tempted to give up, but I think they are really simple machines and solution should not be far from here... Ok I should still consider a completely damaged PCB, but test with ISS ROM says that after all the three main components are working and symptoms are commonly traceable to RAM (checkered screen). So... It is just a matter of time, patience and passion. I only lack the first.

It is time for me to have some rest. I will be on holiday for a week (Dolomiti are waiting for me), so expect no update. As I return I will start again to bother you with questions, my friends. :D

rh74
2nd Star Corporal
Posts: 20
Joined: Mon May 17, 2021 8:53 am

Re: A problematic Oric 1

Post by rh74 »

Hi all, I'm back from holidays... :(

Ok as I told you I gone through all the soldering removing some cold ones and cleaning/soldering other ones. Anyway no change to the Oric. I still have some 170/180 ohm on ground rail and consequently the same RAM/video problem. In the search for a solution, I'm going to restore the TR-1 I removed due to a broken leg. I know I can use a BC547 instead of BC184 but I'm not sure about pins mapping. The BC184 I removed seems to be B-E-C, the 547 instead is a B-C-E. Do you have any mapping advice on that? From the scheme (issue 7) I'm not sure about the emitter (to the ground) and on the collector (to CB1).

Then... How to find a shorted component? Should I cut the common ground track and test block per block? During the my last time assessment I perceived a little current flow on my hand skin, on the left side of the PCB. What could it be?

Thanks as usual!

Post Reply