NEC NC900 failure

[Carsten Kurz]

Hi, a colleague just called me about his NC900 issue that came up today. Until yesterday, everything was working okay with his Doremi DCP2K4 and NC900. The system is about 10 years old. Today, he started up the combo and received these errors. Reboots don't cure it. Do these errors tend to indicate an enigma defect, or could it be the notorious router? They are not at all tech savvy there, although I could probably talk them through a card reseat and remarriage.

Unfortunately, he now also tells me that something like 'Certificate or key error' and 'tamper error' turns up. The typical 4xx errors won't be on the display, only within the NEC or TI S2 software?

IMG_2437.jpg IMG_2447.jpg IMG_2440.jpg IMG_2446.jpg IMG_2445.jpg

[Ioannis Syrogiannis]

Does that colleague of yours have the DCC software on a computer?
It would save them and you effort on pinpointing the problem.
The 400 (Enigma Comm Fail) error has the following troubleshooting guidelines:
Check the connection with NC-80LB, LAN I/F, and firmware version.
Check points:
1 Check the cable connected to the router.
Check whether the power and LAN cables are correctly connected.
2 Confirm that the CPU PWB is firmly inserted in the MOTHER PWB.
3 Confirm that the NC-80LB is firmly inserted in the MOTHER PWB.
4 Confirm that the Enigma board is correctly mounted on the NC-80LB
After conforming the above mentioned conditions, turn the power supply ON and examine the error-related status. If there is still an error, replace the Enigma and NC-80LB PWBs in this order. Turn the power supply ON each time a PWB is replaced. The faulty PWB can be identified according to the error status.

Edit: DCC, not DSS...

[Carsten Kurz]

No, they don't have a computer on site, nor do they have the DCC S2 or TI software, although we could probably arrange for that. The owner does of course have a modern WIN11 notebook. Does the TI software still work with WIN11?

[Ioannis Syrogiannis]

I have zero experience in Win11, so maybe someone else will be more helpful, but I see no reason for it not to be. I can't imagine what would be the breaking point between Win10 and Win11 on such programs that work much as displays/renderers and less for computing info.
I should remind you that NEC S2 projectors have the WebUI as well. An NC900 would have a field with the errors in the „status“ part.
image.png

[Steve Guttag]

Without looking too far into things, I would definitely suspect router since so many things are not communicating. If the Cinema Controller (or whatever NEC is calling the equivalent) can't communicate with the Legacy board, which as the Enigma, then you will get the rest of the problems. The only certificates of interest are the Enigma and ICP. However, without communication with the Engima, it can satisfy this error.

i would start by reseating the cards in the card cage to see if that starts communication again. You will need the user/pw to clear the marriage tamper. I think you can put in the user/pw on the keypad but it won't be pleasant. I haven't done that...always with the DCC for the marriage and, by DCI rules, the keypad to "close" the security door.

[Carsten Kurz]

He can still control the projector from the Doremi - would that still be possible with a broken Router? How difficult is it to replace the router? This NC900 is from 2013, has been running multiple times a week since then, but no part except for lamps and air filters have ever been replaced since the installation. Unfortunately, it's a 350km drive from here, and I am currently too busy to go there.

[Steve Guttag]

That is on the WAN side of things so the CPU (NEC's CCB) board might be able to talk to the outside but not to any of the LAN devices. I'd still start with reseating all three boards in that card cage (CPU, ICP and Legacy). This would likely also relieve any I²C buss issues too.

Though more rare, I have had to reseat an Enigma, in some instances too. Where I see my "reseat" instances go up are when the the RH of the projector go towards the extremes (read dry in the winter months and real high humidity towards the summer months). It's the transition that seems to be where it is worst. Once we are fully immersed, things tend to settle down.

[Leo Enticknap]

If the Cinema Controller (or whatever NEC is calling the equivalent)...

Per the slides from the training, some of the CCB equivalent functions are on the CPU "PWB" (printed wire board - NEC Jinglish for board), others on the slave board, which, IIRC, sits on top of the card cage. Power for the router comes from the slave board.

Agreed with Steve that the router is the most likely culprit, because (a) the failure mode in which some functions fail, but not others, is common, and (b) after the ballasts, it's about the most frequent part in an NC900 (or any Series 2 NEC, for that matter) to fail.

The good news is that unlike the ballasts, the router board is an easy swapout. IIRC, the only panel you need to remove is the lid, and it's immediately visible to the left of the card cage, mounted on a holder vertically. The one gotcha is that you'll need a long (as in, at least 8") no. 2 Philips, ideally with a magnetized head, to get the screws out that secure the holder to the projector's chassis. But other than that, the router card is about the only major internal part in an NC900 that you can swap out without acquiring any gray hairs in the process. It's not that pricey, either.

[Mark Gulbrandsen]

I've had to replace a few NEC routers in the past, and I don't remember there being any errors other than a Com Error on the display. But it may also depend on which channel of the router has died.

[Steve Guttag]

With NEC repairs, don't forget the "cuts" from all of the sharpened metal inside.

[Harold Hallikainen]

Out of curiosity, I just ran this search: https://www.film-tech.com/vbb/search?q=reseat . I get 93 posts with the word "reseat." Are the board connectors gold? Do they have multiple contacts per pin? Why is the solution to a problem so often "reseat the board." Is this a design defect?

[Mark Gulbrandsen]

Yes, they're gold contacts. However, it's not just NEC projectors that have the problem it's all S2 projectors, but mainly NEC in humid areas. Not going to say about S-1 because, thankfully, I only installed a couple dozen of them. But even among the 300+ NEC S2's I installed, I only had this happen perhaps a dozen times. I will also include S2 Christie and Barco, as I had it happen with those as well. All my installs were in States with very low humidity, which does seem to help. I know another Tech that posts here and lives in much higher humidity conditions will say he had thousands of NEC's that neeed reseating and contact treatment. After it happened to me I think 4 or 5 times, I treated all new NEC's with deoxit during install.
I don't remember any multiple type connections, I only remember separate pins and the connectors on the back of the boards were almost as wide as the PCB.

[Steve Guttag]

By far, the ICP module, which has a lot of very low-current contacts (both sides) is the primary culprit. Some projectors have fared better than others. Whatever Barco uses on the signal backplane has the worst record for receiving gold-finger boards. Seriously, it is something like 50:1 Barco reseats to other brands, on the same board. Since it is partially projector dependent, it could be the contact material used on the socket too as well as whatever interaction there is that could cause a barrier to build up.

As mentioned, I've had the Enigma also need a reseat and it uses a different, but also, friction type connector (not gold fingers though). That is more rare. I've had, maybe, less than 5 and probably closer to 2 or 3 units need to be reseated. I'm well over 50% of my original S2 Barco projectors that have had the signal backplane changed due to flaky connectivity. At some sites, I'm up to 100%. NEC and Christie are about the same on the reseat issue. It is always at the transition from high to low humidity or low to high humidity.

[Randy Stankey]

Why is the solution to a problem so often "reseat the board." Is this a design defect?

One hundred percent, no doubt in my mind, it is poor design at the root of this problem. It has been a known problem since, at least, the 1970s and has been well studied since the 80's, yet people still insist upon using card-edge connectors.

[Frank Cox]

Edge connectors can be a good and quite reasonable solution for certain applications; they were originally designed for expansion cards in computers and in that kind of situation they're wonderful. Need a new video card? Rip the old one out, slam in a new one, done. Beats the socks off of having to crank up the soldering iron.

Where this seems to fall down is when designers get carried away and put eight bazillion pins on the card so if it moves a millimeter it's out of position and stops working.

Nothing wrong with the edge connector as a concept, though.