New Member and New Oric Atmos Owner

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.
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Hi.

First I want to introduce myself. I'm James, and I've just become the proud owner of a lovely Oric Atmos. I have experience with retro computers, being a BBC Micro, BBC Master, and Spectrum+ owner, all of which I've nursed back to health.

I bought the Atmos last week, in a faulty state (as it seems the only way to get one at a reasonable price these days), the description saying it turned on, a humming sound could be heard, but the previous owner was unable to tune the TV into a picture.

I too could not get a picture, but typing "PING", "ZAP" & "EXPLODE" worked! a good sign. I received a Scart cable from Retro Computer Shack and I do get a picture from the RGB output, but its showing only 315 bytes free, and there are random "$" appearing on the screen. The keyboard works fine, and I've run a 10 print... 20 goto 10 program with no problems.

In terms of diagnosis, I've had a look through the service manual. The manual mentions a 7905 regulator, but mine has a TDB2905-SP Which is a compatible part, I believe, and mine is reading 8.79v on in the input and -4.85 on the output. This is from a brand new regulated switching supply (the Atmos didn't come with an original PSU). I'm seeing 4.84v on the feed into the modulator.

From what I can see, bearing in mind I have zero experience with these machines, is that the CPU, VIA, AY, ROM, and ULA is fine. Given that it also starts up, ok, I'd say the ram is fine too. but im not 100% given its only showing 315 bytes free.

I have suitable replacement capacitors but im not sure how important the caps are on this board, with the exception of the 100uf cap (C8) next to the regulator. I was thinking to swap that out and see if it made a difference. I was also wondering if anyone had swapped in a Traco TSN1-2450 DC/DC converter? I put the TSR1-2450 in my spectrum.

Now, here's the scary bit. When testing the regulator, I accidentally shorted across the input and output. The speaker went off, so I pulled the plug and reinserted, and now the screen is yellow, but everything else seems to work as before!!

Any suggestions would be gratefully received, Thanks.
Attachments
IMG_0070.JPEG
IMG_0073.jpg
IMG_0074 - 01.jpg
Last edited by SpaceFlightOrange on Wed Nov 06, 2019 1:26 am, edited 1 time in total.
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Ok bad news

I've just turned it on again, and now I'm getting this!!!
IMG_0075.jpg
I think shorting must have killed something!!
robcook
Flying Officer
Posts: 142
Joined: Fri Aug 10, 2012 11:20 pm

Re: New Member and New Oric Atmos Owner

Post by robcook »

On the plus side, that is the coolest looking broken Atmos screen I've ever seen. Unique for me. So thanks. It actually looks like a game prototype or something. Perhaps you should leave the Atmos busted to retain such a beautiful, broken crash screen.

Alternatively, someone here will advise on how to fix the thing as I have no useful advice in that area.

ETA: ¨315 BYTES FREE" - I'm sorry, but I laffed :( I'll go now. Blame Heineken.
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Further update. I'm working through the repair manual, hoping to avoid cutting tracks given it was working earlier. I've pulled the ULA and was still getting 4.85v from the regulator, lifting the OUT leg, now gives me 4.92v across the regulator. I'm thinking to replace the regulator anyway, but something is pulling it down slightly.

I'm going to stop there for tonight, but Any suggestions would be great. I'm feeling more confident handling the board now that its faulty. I think before, I was using kid gloves and my timidness resulted in the accident.

I should mention that I do have a scope, so I can carry out the necessary tests, and I have a spare ULA that arrived today from Bulgaria. I don't have any RAM chips (MT4264-15)
User avatar
Chema
Game master
Posts: 3014
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: New Member and New Oric Atmos Owner

Post by Chema »

Not an electronics expert at all, but you get a stable image (and some words a few days back) so I'd bet the ULA, ROM and CPU are ok.

You had sound, so the AY seems ok too and probably the VIA.

I think the problem is, as usual, the memory. Did you gently touch the chips to see if any gets really hot?

With an oscilloscope you can try to see if there is activity in the address and data bus when booting. An address or data line stuck to low or high is a good indication of a faulty chip. Remember they are 64kx1bit so each chip stores one bit of each byte in RAM.

I can't remember, but probably 4.85 is not an uncommon value... you can also try to measure the current being drawn, but that requires extra equipment. Should be below 600 mA, which is the amperage of the original PSU.
User avatar
mikeb
Flight Lieutenant
Posts: 282
Joined: Wed Sep 05, 2018 8:03 pm
Location: West Midlands, UK
Contact:

Re: New Member and New Oric Atmos Owner

Post by mikeb »

Your non-working boot screen looks like there may be a problem with memory -- if you stare at it long enough you can sort of see the "correct" black and white bar pattern in there ...

I must say, shorting the input to output of the regulator is quite a bad idea (this would put 8-9V onto TTL circuitry, which is very intolerant of overvoltage), and I'm amazed you haven't killed something outright. It may be that IC22 (74LS365) is not working right, if your RGB output has a funny cast to it, but beware, you may have accidentally ended up with YELLOW paper/BLUE ink due to memory corruption -- if the problem persists, once you get things working, try issuing in turn PAPER 0, PAPER 1 up to 7 to see if you get BLACK, RED, GREEN, YELLOW, BLUE, MAGENTA, CYAN and WHITE backgrounds in that order.

You mention a service manual, is that the official Oric one? If so, also check http://oric.signal11.org.uk/html/repairguide.htm for some tips of things to check in a more sensible order.

The random dollar signs: These are #024 (hex) and not #020 (space), so it looks like you might have (had) a problem specifically on D2 (Data bus line 2) which relates to one particular DRAM chip (IC17) -- if you were going to target a chip, that would be a good start point. Or, in fact, any DRAM chip of the 8 that is noticeable HOTTER than the others when it's been on for a short time. It's not hard to spot with a calibrated finger, if one is unhappy :(
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

mikeb wrote: Wed Nov 06, 2019 6:31 pm Your non-working boot screen looks like there may be a problem with memory -- if you stare at it long enough you can sort of see the "correct" black and white bar pattern in there ...
Yes, I had noticed that, could that mean the ROM is probably ok?
mikeb wrote: Wed Nov 06, 2019 6:31 pm I must say, shorting the input to output of the regulator is quite a bad idea (this would put 8-9V onto TTL circuitry, which is very intolerant of overvoltage), and I'm amazed you haven't killed something outright.
I'm fully aware its a bad idea and it certainly wasn't on purpose. Believe it or not, I was using insulated probes! yet it still happened!

I have the Official Oric one and also Mike Brown's It's actually Mike's im using right now.

Thanks for your advice. I'm really quite upset about what I did to it, but as you say, the ram was most probably faulty anyway. Do you have any suggestions on where to get memory from (MT4264-15)? I've seen batches for sale on eBay, shipped from China. I'm a little bit wary of them but they were so cheap I though it was worth a punt.
User avatar
Dbug
Site Admin
Posts: 4444
Joined: Fri Jan 06, 2006 10:00 pm
Location: Oslo, Norway
Contact:

Re: New Member and New Oric Atmos Owner

Post by Dbug »

SpaceFlightOrange wrote: Thu Nov 07, 2019 10:43 pm
mikeb wrote: Wed Nov 06, 2019 6:31 pm I must say, shorting the input to output of the regulator is quite a bad idea (this would put 8-9V onto TTL circuitry, which is very intolerant of overvoltage), and I'm amazed you haven't killed something outright.
I have the Official Oric one and also Mike Brown's It's actually Mike's im using right now.
Hmm, I'm wondering where this Mike Brown is, maybe he could provide some help ;)
User avatar
mikeb
Flight Lieutenant
Posts: 282
Joined: Wed Sep 05, 2018 8:03 pm
Location: West Midlands, UK
Contact:

Re: New Member and New Oric Atmos Owner

Post by mikeb »

Dbug wrote: Fri Nov 08, 2019 11:58 am
SpaceFlightOrange wrote: Thu Nov 07, 2019 10:43 pm
mikeb wrote: Wed Nov 06, 2019 6:31 pm I must say, shorting the input to output of the regulator is quite a bad idea (this would put 8-9V onto TTL circuitry, which is very intolerant of overvoltage), and I'm amazed you haven't killed something outright.
I have the Official Oric one and also Mike Brown's It's actually Mike's im using right now.
Hmm, I'm wondering where this Mike Brown is, maybe he could provide some help ;)
Naughty Dbug ;) ... I should have said "self promotional plug follows" ...

I know it's easy to slip with insulated probes, been there done that :)

The MT4264-15's -- well I'd give any 4164 pin-compatible DRAM that's 150ns or faster a go. Slower is not good (although you can get them, avoid them, as Oric runs the DRAM at 3 accesses per clock cycle). There are so many chip numbers that fit this bill as it's a standard part with many other manufacturers.

P.S. The "cutting tracks" that you mention is more in relation to isolating circuitry that's dragging the voltage rails down to the point of non-operation, at 4.8v you're just within, so you may not need to do that (yet!). It's just that isolating parts of the voltage rail is easier, to my mind, then pulling out chips that are soldered in. The intentional damage to the voltage rail is easily bridged/unbridged in future, whereas pulling ICs out of old boards is liable to create a few new problems on the way. If you do pull any chips, use sockets to replace them.
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Dbug wrote: Fri Nov 08, 2019 11:58 am Hmm, I'm wondering where this Mike Brown is, maybe he could provide some help ;)
:oops:

Doh! Sorry Mike! It didn’t register at all! I’m a bit sleep deprived at the moment due to overnight releases (I’m a lead dev) and sickly children
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Hi.

I finally have an update.

While waiting for the RAM to arrive I desoldered all 8 chips and socketed them continuity tested them thoroughly and put the chips in the sockets. When I tested I got exactly the same pattern as before so I assumed that was successful.

The RAM arrived today (MT4264-15) they came from China and took a while.

I swapped out IC17 as that was getting hot and when i connected up i got nothing at all. so I disconnected and discovered that I had plugged the RetroComputerShack RGB cable into the Tape port!! I hope I haven't done any further damage!! It's almost like my subconscious is telling me to kill this machine!!

pulled the plug and put the original IC back in and connected up, still the same output. new IC back in, and then got this:
IMG_0146.jpg
IMG_0146.jpg (254.47 KiB) Viewed 8142 times
There's a lot of flickering on this screen. I videoed it. See Here

I then noticed IC16 was getting hot (it wasn't before I swapped 17) so I swapped that and the pattern was similar.

to sum up, I have now replaced all 8 IC's and I get this.
IMG_0148.jpg
IMG_0148.jpg (267.68 KiB) Viewed 8142 times
I also noticed that the sound from the speaker got louder at this point

Any thoughts? I guess its time to break the scope out. If you could provide some suggestions on where to probe, that would be great.

Thanks
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Also,

I replaced the regulator and the Electrolytic caps before I socketed the RAM.

Thought I'd share some pics so people can see the soldering work.

Before (Note IC12 has already been removed)
IMG_0116.jpg
IMG_0116.jpg (411.51 KiB) Viewed 8138 times
After
IMG_0119.jpg
IMG_0119.jpg (494.51 KiB) Viewed 8138 times
IMG_0117.jpg
IMG_0117.jpg (438.39 KiB) Viewed 8138 times
User avatar
Chema
Game master
Posts: 3014
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: New Member and New Oric Atmos Owner

Post by Chema »

Last repaired oric I heard about was due to a broken multiplexor (the chips over the RAM, 74LS257). It was replaced by a 74HCT157, which is compatible in this case. Maybe it is worth trying...
SpaceFlightOrange
Officer Cadet
Posts: 44
Joined: Sun Nov 03, 2019 9:44 pm

Re: New Member and New Oric Atmos Owner

Post by SpaceFlightOrange »

Chema wrote: Tue Nov 19, 2019 1:52 pm Last repaired oric I heard about was due to a broken multiplexor (the chips over the RAM, 74LS257). It was replaced by a 74HCT157, which is compatible in this case. Maybe it is worth trying...
IC20? they were my next port of call. I don't have any so I was going to ask what would be a suitable replacement.

Thanks.
User avatar
iss
Wing Commander
Posts: 1641
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: New Member and New Oric Atmos Owner

Post by iss »

2 easy things to try before changing LS257 - fill the via with solder and check that flying CAP's pin do not touch underlying IC pin:
IMG_0119.jpg
IMG_0119.jpg (41.03 KiB) Viewed 8122 times
Additionally you can try to temporary remove the capacitor. It's purpose is to delay slightly the RAS signal which is not always needed (i.e. depend on RAM chips used).
Post Reply