What's new
Pinball info

Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

DMD graphics missing some detail - with possible solution

Vince69619

Registered
Joined
Feb 17, 2020
Messages
28
Location
Reading UK
Hi All,

I had an intermittant problem with my DW machine and the DMD. I am still learning how these things work as my background is self taught electronics rather than looking after pinball machines, but this is what I saw and did...

Mostly the graphics were ok, scores and general images seemed good, but some images, especially the people (Ace & Doctor) whould be really rough on the DMD. Ace would look like a skeleton sometimes. But this would only be for short periods and then it would go back again. As a fault finder, I usually take a guess at what this could be and I totally incorrectly guessed that the DMD controller was loading these more detailed game specific images from an eprom on the controller board and there might be a multiplexer with a lazy bit. I couldn't have been more wrong.

I took out the DMD controller, looked at schematics and discovered the bigger chip under the warranty sticker is actually a RAM chip, so there is no ROM on the board. All the data has to come from the CPU board, so I decided to go ahead and re-cap the board as I have ordered a complete set of capacitors for the machine and bingo, C3 which is a 100uF 10V cap was showing as a 25uF capacitor with an ESR off the scale. Capacitors shouldn't be too far (20%) off their stated capacity and should have a low Effective Series Resistance. The HV smoothing caps were fine, but got replaced anyway and since C3 was replaced, no bad graphics. For my initial guessed scenario to have any merit, bad data would have had to be coming over from the CPU board and much more of the graphics would have been corrupted. Not sure why it presented the way it did, but will be checking more of the smoothing caps on the 5v rails.

Hope this helps someone, or starts a conversation.
Vince.
 
Bloody hell Vince “self taught electronics” I’ve been messing with pins for 30yr now and you lost me somewhere around the opening sentence.
My fix would have been replace that board with this new one after trying another DMD and if don’t work cry for help and the troops come running. :rofl:
thank god for pinfo and my goto Chris and Phil
 
Have you run the display test from the menu ?!

If it was a ROM error - you would almost certainly get a start up error.

Would maybe suggest switching ribbon cables first. At least re-seat them first.

Also unplug the sound board and see if that helps.
 
Well I have been "self teaching" myself electronics for about 45 years ;)
Two guys in my area have an Addams each and I have helped them out a few times with different things over a couple of years, so my pinball knowledge is behind my electronics, but I think I am catching on.
I just thought it was worth documenting that strange graphics can be caused by a faulty C3.

I have a more mechanical problem to ask for help with, might invite Mike Parkins over after lockdown as he is local to me, but some of the lamps on the playfield are recessed slightly into the playfield so sometimes a ball will skip in and out of these.

Vince.
 
Hi Pick Holder,

Yes I did run the test and it all seemed fine and reported the RAM test as passed, so no fault found. Of course I had to re-seat the ribbon cables to get the board out of the machine so if it was a bad connection on those I have dislodged it. And I guess from the reference to the sound board, that board is also sitting on the bus?

So absolutely true that I may have made the wrong correlation about the cap being the root cause. At work we try to dispel false pattern matching, i.e. the system runs fine if I put my small furry mascot (not a metafor) on top of the system cabinet and I wouldn't want to be creating false patterns.

The cap definately was past it's prime. it *could* have been a bad connection on the ribbon cable and not noise/dropout on the 5v rail. Is bad ribbon connections a common issue?

I forgot to mention that I did also reflow all the power connections, i.e. the socket strips that are not ribbon cabled.

Vince.
 
Hi All,

C3 was innocent.

My graphics are missing the half tone pixels again. I have reseated the ribbon cables and tried with the sound disconnected. Still the same.

I might meter out the ribbon cables but for now I am leaving the machine running to see if it changes on it's own.

Vince
 
The display did not get any better. I am learning as I go, so I am now thinking that I need to check the voltage rails. Looking online I notice references to LED DMDs as an upgrade, and with the high voltages that the traditional DMD uses, I am concluding that the display must be some sort of late version of a discharge tube. Missing the dimmer pixels could be because one of the power rails is wrong.

I will let you know.

Vince.
 
The -125v rail to the DMD is only at -116v. The -113v rail is at -104v, so it is 12v higher than the other rail as expected. Is the -125v rail critical?

The +12v rail is really low at 8.5v that isn't good. Need to look at that also.

Anyone know?

Vince.
 
I have replaced quite a few power diodes and bridge rectifiers to sort out low voltage problems so that is probably your 12v issue, but it's also common to find dry joints and loose connectors causing the same problems. As for the DMD, it is certainly advantageous to go LED (or LCD) if you can bear the cost. Colordmd looks great but is around £400 - I recently took part in a group buy on here for pin2dmd at around £100 each but I don't know if Dr.Who has been designed yet. If not, you can still change the default palette to any 4 colours of your choice.

Hoping to pay a visit one day, Covid19 permitting... 😷
 
Hey @Vince69619 - are you sure the display itself is not gassing out?
Do you have another DMD game you could swap them over ??

Also @Mike Parkins - the next run of Pin2DMDs are going to be over £150 from what was previously posted. The first lot were about £100 due to the bulk of the amount of panels ordered.
 
@Mike Parkins I tracked the 12v to TP1 on the driver board and it is good there, I think it was about 15v unregulated, so it probably is bad connections. It got a bit late last night to go poking around. The AC feeds into the DMD controller are well above spec, so I need to look at the PSU on that board. First suspects are board deposits or zeners. We would love to invite you round, when we are allowed.

@Pick Holder I don't have anything else here, I might be able to swap things over with my mates Addams when the virus is gone. I am tempted to try to get these power rails closer to spec and see if that helps, what do you think?

Thanks for your help guys.
Vince.
 
@Pick Holder - I will test the leads I have and report back. I have noticed a discrepancy on the drawings for this -125v power rail. The Addams schematic I have has been modified (badly) to call for a pair of 56v zeners, and the parts list in the Dr Who operations manual says they should be 1n4758A 62v - which don't exist, the 4758 is 56v, it's the 4759 that is 62v. Twice 56 and a little overhead is what I am getting.

I think I will check the power connections are clean and have a better look at these ribbons.
 
I personally no longer refurb/repair those dmd boards for the simple reason - they are very tired. Well over 25 years.

Also people are buying pin2dmd's or colordmds now. I do not think a colour version of Doctor Who exists at the moment for Pin2dmd.

Can you record a video of what the display is doing and upload it to youtube?
 
What it's doing is simple, it is missing out any low intensity pixels. It's not looking corrupted in particular.
 
That is why I was checking the power rails, in case these had an impact but now I think the high negative rails have been consistent all the time. If it's a ribbon cable, it's only very slightly out.
 
I do not think a colour version of Doctor Who exists at the moment for Pin2dmd.
Yes I think you're right, which surprise me as it would look great. There is a dormant w-i-p and I coloured a few scenes for Solar Sailor but it's really hard without access to the real machine.
 
Hey @Vince69619 - are you sure the display itself is not gassing out?
Do you have another DMD game you could swap them over ??

Also @Mike Parkins - the next run of Pin2DMDs are going to be over £150 from what was previously posted. The first lot were about £100 due to the bulk of the amount of panels ordered.
Or you can built your own pin2dmd if you can solder
 

Attachments

  • IMG_20200416_173529.jpg
    IMG_20200416_173529.jpg
    153.9 KB · Views: 16
@Balazs Oh yes I can solder, I can lay my hands on 4 irons and a hot wand without going far. Do you have a link to the pin2DMDs so I can see what it's all about? -Edit, no need, Google found it!
 
I bought the panels from aliexpress for 36£
Development board from farnell for 20£
And the shield from germany for 10 euros+ postage
 
I bought the panels from aliexpress for 36£
Development board from farnell for 20£
And the shield from germany for 10 euros+ postage

Do you have a list of direct links for those please @Balazs ??

I also seem to remember there is a 3d printable backing frame somewhere...

Cheers
Paul
 
I have worked out what is happening with my 12v supply. I am working off the Addams schematics as my Who schematics are in the USA until lockdown is cleared. On the Addams, R7 on the controller board is a wire link and the zener D8 is not in use, but the Who manual says R7 is a 56 ohm resistor with a 12v zener in D8. R7 is dropping about 5v, so needs more investigation.

Wouldn't the DMD be the same in these two machines?

Pin2DMD does look interesting. I will read more later.

Vince.
 
German Gaming Supplies.... oh dear... He is very unreliable at the best of times.
Pretty well runs a lot of the facebook pages as well about visual pinballs.
Google 'german gaming supplies problems'

I would use this guy in france:
or


Would recommend you get a frame for the panels as well - plus you have to pay a registration fee to use the firmware for pin2dmd so it all adds up. Plus a memory card.... 5v Power supply.... It is also a nicer looking display to buy the EVO boards that was used on the last buy in here, there are MUCH nicer. The older pin2dmd system as well is a lot more untidier, oh and a pain in the back side to get the software into in the first place (once you do one its easy but not much online help).

Back to Dr Who....

If you are missing the low density pixels then it sounds like the DMD to me. If its intermediate as well.... If you PM me I could send you a spare normal DMD.
 
All fixed.

I took the advice and bought a new set of IDC cables for the machine.

I need to back track or test the older cables in order to work out which one caused the issue, and I do plan to do that. (I might make an arduino shield to test multiway cables, I might just put the old ones back one at a time and try to break it.)

If I work out which one caused this, I will update this thread.

Vince.
 
I need to back track or test the older cables in order to work out which one caused the issue, and I do plan to do that. (I might make an arduino shield to test multiway cables, I might just put the old ones back one at a time and try to break it.)

I wouldn't bother TBH, the cable sets are cheap, they are well known for giving troubles and strange symptoms, just bin the lot would be my advice and enjoy the knowledge of not trying to reuse and old cable will cause you further grief in the future.

Chris.
 
Back
Top Bottom