NC2000 LPSU Problem

[Leo Enticknap]

Wow! Someone actually got an error message from a projector that accurately described what the problem was? Well bugger me with a soldering iron! I never thought that could happen...

[Marcel Birgelen]

I think you can safely replace "projector" with "random thingamabob".

But there is a deeper problem with error messages and failure modes those days: Things aren't designed to fail, that's why most error messages are cryptic; this isn't supposed to happen, so I throw some non-descriptive error message and let somebody else figure it out. If things were designed with the possibility of failure in mind, they could, in many cases, work around some of the failures all together, or at least guide the system operators in such way that full functionality can be quickly restored.

[Bruce Cloutier]

We could examine error reporting and the use of assertions, exceptions, errors, warnings, etc. as they relate to the end user at length and not change the world. I code a lot of such things. I've been including the line number from the source code in the error so I can figure out what happened no matter what the text of the message says. And, in this case I know that I will be the one at some point that will be responsible for the resolution.

But, I think Leo references the frustration that all of the technological complexity has brought upon us with a, perhaps well-stated, bit of levity.

[Marcel Birgelen]

We could examine error reporting and the use of assertions, exceptions, errors, warnings, etc. as they relate to the end user at length and not change the world. I code a lot of such things. I've been including the line number from the source code in the error so I can figure out what happened no matter what the text of the message says. And, in this case I know that I will be the one at some point that will be responsible for the resolution.

But, I think Leo references the frustration that all of the technological complexity has brought upon us with a, perhaps well-stated, bit of levity.

Maybe I'm getting a bit too pseudo-philosophical here, but since we're living in interesting times, maybe there's a bit more room for this.

I've been writing all kinds of code for over 30 years now. I've been debugging and "repairing" stuff since I was like 6 years old. The world has indeed become quite a complex place to live and advanced technology has penetrated every fiber of our life and it's continuing to do so.

But, you can probably also see that the way we approach technology is fundamentally wrong.

For example, I've been debugging some kind of stupid web-app that needs some "COVID-19" updates for the better part of the day now. The code ran on a machine that hasn't seen an update since 2014 and I decided to bump it up to 2020, since the whole thing was a bomb waiting to happen. The thing is written in Django, which is a very popular web-framework written in Python.

Now, since 2014 and 2020, almost all libraries and sub-framework components have been completely changed... Some libraries have disapeared, others have seen their API changed a little, others have grown into unrecognizable monstrosities, not even resembling their former selves.

Exceptions thrown in stack traces literally 200 levels deep often make no sense whatsoever, as I don't have a clue what all those layers in between are doing and neither should I actually have to care.

So, to get this code to a "supported platform", it looks like it's going to need an almost complete rewrite. This isn't code that's 30 years old and written in an obscure language only a select few remaining breathing relics are allowed to touch, it's code that's barely 6 years old and it's literally worthless now.

Over the years I've come to the conclusion that how we handle technology is fundamentally flawed. The way we develop hardware is fundamentally flawed and the way we write software and build systems is also fundamentally flawed. That's why everything around us is constantly in a state of breaking down. Maybe it's the fringe in which we humans want to live, maybe we can only really function in a world where everything is a few seconds away from going haywire. The irony is, we could do so much better, if we only even tried...

Most code I've read in my life is garbage and to be honest, most code I've written is, as a result of the ecosystem it needs to operate in, also garbage. We can design systems that can potentially run forever. We can design systems that, when they fail and can't restore themselves, we can at least tell the users or the operators what actually went wrong, instead of some cryptic error message. But we're not doing so, because everything is driven by short-term values.

[Torsten Juergens]

Hi,

unfortunately I also have an acute LPSU problem with my NC2000C. At the end of the show the lamp went out suddenly.
Event Log:
2020/10/11 19:45:03 (125) LPSU Fail (Power)
then again and again
2020/10/11 19:50:17 (125) LPSU Fail (No Lamp)

I changed the lamp, removed the LPSU, opened it and reseated all plugs. The two fuses of the LPSU checked, both ok.
After re-installation the error continues

2020/10/11 19:50:17 (125) LPSU Fail (No Lamp)

What is left besides replacing the LPSU?

Greeting
Torsten

[Leo Enticknap]

I'm afraid it looks like you need a new LPSU. From the online service manual:

error_125.PNG
Sorry.

[Torsten Juergens]

Ok, but you don't want it to be true.
Thanks Leo.

[Leo Enticknap]

It's a pricey part. That's why I didn't want it to be true.

[Torsten Juergens]

Fortunately, I was able to repair the KSX-4001MPNE LPSU.
I built in the two power supplies from a used KSX-4000NEVT.
And since then everything works again

[Leo Enticknap]

So either the transformer or the AC to DC conversion circuitry had gone bad. Glad you managed to dodge the bullet of having to buy one of the more expensive NEC parts.