24" Widescreen CRT (FW900) From Ebay arrived,Comments.

Have a look at the 2 2200µF/10V on the G board, I've seen them drop pretty much in capacitance on two screens (even below the 20% tolerance). The 1000µF/25V or 16V also tend to lose capacitance on the D and N boards though not as much. Same for the 220µF/10V or 16V (C684, C516, C909, C437). The 47µF/10V or 16V seem to take some punishment on the G board but that's seen at the ESR level. 47µF/25V may have increased ESR depending on areas.

In general the electrolytic capacitors in the worst shape are the decoupling ones on the current supply lines with voltage ratings of 25V or below.
 
My usual 'net lurking turned up this Reddit thread on FW900 repair:

Symptoms sounded similar enough, so I tried firing up mine again just to check. Solid green light, relay thunk, spark gap bulb lights and buzzes for about a third of a second, monitor quickly shuts down, green light goes solid amber, doesn't seem to restart.

One of the posters suggested checking MOSFETs Q901, Q905 and Q906 to start with. That could give me a lead into how to resurrect this thing inexpensively instead of letting it collect dust like it has for the last two years, though I'm reasonably sure the issues with bright color flashing upon cold startup when it did work are rooted much more deeply that that.
 
Oh, interesting piece of information about the mosfets.

I'm trying to fix some horizontal interferences on the GDM5410 I talked about previously (a couple of lines slightly vibrating here and there). The issue seems to vary a bit depending on if the screen is cold or not. I tried changing quite some high voltage ceramic capacitors everywhere but even though some were a bit tired, that didn't fix the issue. All electrolytics are new. So unlikely a filtering problem.

Some information here: http://www.repairfaq.org/sam/monfaq.htm#monrwphb
They talk about issues on the B+ voltage line (220V in our case, which feeds the horizontal sweep, G2 and flyback). For some time that screen had SG901 often activated and it finally stopped by itself with use, that would support an issue with the voltage line feeding the flyback. I tested most of the passive components around that line, they seem OK. So in the end it could well be the mosfets aging and jittering, especially if it's a known weak point ...

Now the real problem is to get a suitable replacement part if it's needed. The one in Q905/Q508 has a fast response time to open but is much slower to close, I guess the regulation system relies on this to make the right output voltage. I've not found any recent mosfet with such characteristics, and anyway depending on the model the test conditions to obtain technical values varies, it's a bit tricky to compare.
The safest shot would be the original part (the one which fails ...), but good luck finding a reliable source for such obsolete parts. :unsure:
 
People in the US are lucky with so many of those monitors around. That's almost impossible to find in France.

Little update with Sherlock Holmes' quest to find the root of issues ... Desoldered Q905, Q906, Q508, functionnal and the few values given by the tester are within specs (except a capacitance > 2 nF while the 3 ones listed on the datasheet are in the 50-1000pF range, not sure what this measurement is really related to). That doesn't entirely remove suspicions though. Put some fresh thermal paste and resoldered, still some vibrations in the display. Still searching ...

By the way if people had to change Q906 or Q901 (Fuji 2SK2655-01R), I found a recent mosfet which seems very similar in specs: Fairchild FQP9N90C. And it's much easier to find than the original of course
 
Last edited:
I've got an SGI branded FW900 i acquired from ebay a few years ago. When you turn it on cold, you get the usual diagnostic information (Monitor is working) followed by some quick green flashes with zigzags going across. A few times the monitor has shut itself down (i guess when its so cold). I'd say most of the time the image will stabalize, although with a bright tint for the first 30mins of use. After that its all good.

Today i opened it up and had a look inside. I removed the power board and inspected it for any dry joints. I also looked at the other boards as best as i could while in situ. No signs of bad caps or anything.

I do have another perfectly working FW900, so this one to me is kinda surplus. It's not in fantastic shape, as the previous owner did not mention to me any of the two tiny scratches or the condition of the front bezel. I believe the screen coating has been removed. Two bottom screws from the outer case are missing and there was some what i believe to be dried pellets of pet food inside!

I have good experience with electronics, but CRT's are not my forté. I am debating whether to let this one go to someone else. Surely there must be a flat panel that is capable of reproducing the same level of colours by now???

My SGI branded FW900 has started to do this too, I turn it off quickly when it flashes green then switch it back on and seems to be fine.

It has happened more since I moved it to a colder room. Should we be worried?
 
People in the US are lucky with so many of those monitors around. That's almost impossible to find in France.

Little update with Sherlock Holmes' quest to find the root of issues ... Desoldered Q905, Q906, Q508, functionnal and the few values given by the tester are within specs (except a capacitance > 2 nF while the 3 ones listed on the datasheet are in the 50-1000pF range, not sure what this measurement is really related to). That doesn't entirely remove suspicions though. Put some fresh thermal paste and resoldered, still some vibrations in the display. Still searching ...

By the way if people had to change Q906 or Q901 (Fuji 2SK2655-01R), I found a recent mosfet which seems very similar in specs: Fairchild FQP9N90C. And it's much easier to find than the original of course

Is this detective work with respect to that red issue you're having with the 21 inch Sun GDM-5410?
 
Is this detective work with respect to that red issue you're having with the 21 inch Sun GDM-5410?
The color issue is fixed. ;)

That's another issue with the same screen. There are a few horizontal lines vibrating here and there in the display. Actually not that disturbing on this screen but I've seen the same thing on some others, including a FW900. It's worth finding how to fix that.
 
ah well done! Sorry if I missed a post where you already said that.

Are the wobbling lines usually the same ones each time? Or is it random?
 
I've drawn marks with a paper pen on the bezel to check if they are still there or not after each change. It always appears at the same places for a given resolution/frequency. But the intensity of the phenomenon seems to vary depending on the duration the screen has been on (or its temperature).
 
I use a microfiber cloth combined with an anti static screen cleaner. But the micro fiber cloth alone doesn't work super well. Might try the ones you linked to, thanks.
 
i got them for $15 so the price might drop :p

my thoughts on cleaning:
1. if the ag film has been removed, it doesn't really matter what you do since the glass is tough and stuff doesn't show up as obviously as on the ag film
2. some things which appear benign actually leave residue on the surface: canned "air" and tape. learned this in undergrad while trying to clean silicon wafers
the residue may be invisible, but in the long term it may cause more dust to adhere
3. cleaning should be done in 2 steps: first remove large dust particles as gently as possible to avoid scratching the screen and then remove any oil or whatever on the screen

currently mine still has stuff like this
http://i.imgur.com/4tHeeuZ.jpg

in other news:
did wpb on this monitor yesterday. last wpb i did was in august 2015
optimal g2 has drifted from ~164 to ~159 or so
c_max_b_max on the 2nd step is 105
 
in other news:
did wpb on this monitor yesterday. last wpb i did was in august 2015
optimal g2 has drifted from ~164 to ~159 or so
c_max_b_max on the 2nd step is 105

That's not a lot of black level drift. Nice :)

Also, had a chance to do some experimentation with a CRT tester a while back. Have learned some stuff, but there are a few more wrinkles I need to sort out. Anyway, one of the things that emerged is that I'm pretty sure a multimeter can be used on the CRT pins to figure out what each step in WinDAS is actually doing.
 
I've just given a try to the Vivanco adapter we talked about previously, it's really disappointing. Unless I got a bad unit I think it can be forgotten.

The good news is that I managed to push it up to 296Mhz (1920x1200@89hz using the standard blanking settings for the FW900).

The bad news is that as soon as you start pushing it a little (I'd say, above 70-100 Mhz maybe ?), you get a bright trail on the right side of any character and the picture seems to become also quite blurry. I tried with 2 different VGA cables, 2 different screens and 2 different computers. The adapter is definitively the issue.
 
thanks for this info Strat. Disappointing! Did you contact the company to ask them if this behaviour is expected?
 
Nope. But since there was a guy using this adapter who described the same bright trail issue on another forum, I'm afraid it is a "feature". It's surprising that he also talked about crystal clear picture, it's not really consistent with such kind of problem.
 
Is it OK to run CRTs on usa 60hz 240v 2phase if the monitor is rated for 100-250v? Specs on back say nothing about neutral requirement.

Thinking of running a 240v line to PC to increase efficiency and use up some extra copper.
 
That's not a lot of black level drift. Nice :)

how much do yours drift in a year?

i just redid wpb using g2=158 instead; contrast ratio went from ~3500:1 to ~10000:1.
edit: checked the saved windas file. previous g2 was 163. so in roughly 1.6 years, the g2 necessary to get ~10000:1 contrast ratio went down by 5.
cspmoPB.png

close enuf

maybe we've discussed this before, but apparently the i1 display pro doesn't lie flush against the screen due to the position of the cable. any good solutions?
currently i'm balancing my wallet on the top edge of the i1dp. maybe one way would be to add something to the top of the screen bezel, such that the i1dp's cable runs parallel to the monitor.

also: i feel like it's been a few years, but is i1dp still the best for its price?
 
Last edited:
oh
and during the wpb, i randomly saw some 2 or 3 white flashes coming through (i think) the back of the crt

did something in there zap a bug perhaps?

the crt image wasn't affected when it happened but i saw a bit of the white flash through the borders of the screen
 
how much do yours drift in a year?

That's something I should pay attention to next time. I can't really remember. My monitor is on a lot though - sometimes 16 hours a day. And I do set the G2 such that the blacks are super inky, but I don't think this would increase the rate of whatever phenomenon causes drift in the first place.

maybe we've discussed this before, but apparently the i1 display pro doesn't lie flush against the screen due to the position of the cable. any good solutions?

I lent my colorimeter to a friend, so I can't test now. But I'm pretty sure I was able to get it flush. I think in order to do that I had the cable at a slight angle (so it wasn't perfectly vertical). And I also used a book or something heavy to pin the cable down to the top edge of the monitor. Wish I could remember exactly how I did this. I'll take a photo of how I do it when I get it back from him.

also: i feel like it's been a few years, but is i1dp still the best for its price?

I just browsed a few recent threads on avsforum, i1d3 still comes highly recommended. I didn't see any indications of something of better value. My take on this instrument is that it really is an outstanding piece of technology.

oh
and during the wpb, i randomly saw some 2 or 3 white flashes coming through (i think) the back of the crt

did something in there zap a bug perhaps?

the crt image wasn't affected when it happened but i saw a bit of the white flash through the borders of the screen

heh, never experienced that, but that's a pretty glorious death for a bug... death by CRT :)
 
Bad news,Simon has replied by saying that the manufacturer has changed idea due to VGA resolution is not required beyond 1080P according to mass feedback.
They say that there isn't mass market appeal.
Very strange,they make a prototipe,expect CE approvals and then nothing,they could say it before and not wait six mounths.
I'm sure there are adapters with the right chip out there,the problem is find them

To Strat_84
Have you tried to contact Vivanco to know if that problem is normal?
Maybe it can be resolved with a different firmware,that thing can do near 300 MHz
 
Bad news,Simon has replied by saying that the manufacturer has changed idea due to VGA resolution is not required beyond 1080P according to mass feedback.
They say that there isn't mass market appeal.
Very strange,they make a prototipe,expect CE approvals and then nothing,they could say it before and not wait six mounths.
I'm sure there are adapters with the right chip out there,the problem is find them

???

Have they seen this thread? There may not be a ton of people posting in this thread, but I feel it represents a larger community of people that have held on to their CRT's.
 
As far as I understood, their adapter was the regular one, with only a custom firmware? Just like this very cheap one?

http://www.ebay.com/itm/DisplayPort...A-Adapter-Converter-Gold-Plated-/152441491686

And there is only one out there, the one amlett has, that works all the way up to 1920x1080@96Hz and 2304x1440@72Hz? And that, just because of a firmware upgrade?

If only we could simply get the regular adapter and upgrade the firmware... :(

Right now, we are stuck with nothing planned at all, with Analogix who clealry won't do a thing for us, and HDFury struggling to keep selling their products. They say they made a lot of progress on the design, but now I'm not sure they will ever be able to release the HDFury5...
 
As an eBay Associate, HardForum may earn from qualifying purchases.
Well, I went back in this thread, and I saw that.

I sent a query to analogix last thursday and they call me on friday. I tell them that their motherboard chipsets ANX6212 support 2560x1600@60Hz, 24bpp but the ANX9833 external seems limited in their specs to 1920x1200@60Hz, 24bpp.

The guy who call me told me that both use the same chipset, so he'd clarify why the externa one had different spec on the website.

So I downloaded the product briefs of both chipsets to see the differences. Both are 48 QFN ICs, with the exact same bloc diagram, using the same power supply and with the same power consumption. Sadly, I can't find the pinout, and that would be really helpful.

Anyway, both chipsets might as well be the same, with simply a different firmware, I don't know. It could be interesting to get an adapter, and the motherboard chipset, and swap the chipsets, see what happens... QFN parts are not exactly easy to desolder though.
 
ANX9833 has a DP receiver that works in HBR mode (2.7Gbps per lane)(max bandwidth 180 MHz)
ANX6212 has a DP receiver that works in HBR2 mode (5.4Gbps per lane)(max bandwidth 360 MHz)
Maybe the DAC is the same,maybe the DP receiver is the same but with different firmware that enable the HBR2 mode
I had the same idea about swapping the chip,both the adapter with ANX9833 (i have one) and the chip ANX6212 are on the market,but i don't have the equipment for the work
 
I just bought two of the eBay adapters I linked above, hope these are okay. I have basic equipment including a (cheap chinese) hot air rework gun, since I work with electronics as a hobby. I know for a fact I don't work well at all with parts smaller than TSSOP, that includes QFN here.

I tried to find the pinout of both ICs, but no luck. I just sent an email to Analogix, maybe they can help me with that, but I'm not sure. I mean, they manufacture the chips, not the circuits that uses them, right? So they should have all the informations needed for the integrators to build their own circuits. I'm not sure why I can't find a more detailled datasheet. With the proper datasheet I could even try to design an entire DP to VGA adapter circuit. I'll keep looking.

Can you please let me know where you bought the ANX6212 chips? I've been looking for these for the past few hours, I can't find them anywere for sale!

EDIT: Forget about the last thing I said. I just found and bought 5x ANX6212 ICs from Aliexpress.
 
Last edited:
Bad news,Simon has replied by saying that the manufacturer has changed idea due to VGA resolution is not required beyond 1080P according to mass feedback.
They say that there isn't mass market appeal.
Very strange,they make a prototipe,expect CE approvals and then nothing,they could say it before and not wait six mounths.
I'm sure there are adapters with the right chip out there,the problem is find them

To Strat_84
Have you tried to contact Vivanco to know if that problem is normal?
Maybe it can be resolved with a different firmware,that thing can do near 300 MHz

Ah ah, mass feedback, nice one. Like they would have lots of people calling them everyday for VGA adapters, but only for shit resolutions. That makes no sense. And even then, the best can do the worse, it's only a firmware update, not like it would cost a lot to manufacture improved ones ...

I have tried to contact Vivanco, just in case. Their UK support told me they forwarded my feedback to the HQ in Germany and asked them to contact me directly. That was one week ago, no news. My guess is that they know their product isn't exactly of the best quality and they don't really give a shit about it.

I'm also starting to consider if handmade DP -> VGA converters wouldn't be the only way to have proper performance and quality.


etienne51 : Be careful with Aliexpress, lots of counterfeits or used/damaged electronics there. That's a source I use only as a last resort and with a lot of defiance.
 
Yes, I also don't really appreciate Aliexpress myself. But I figured, since it does not cost much at all, let's give it a try, I couldn't find the part anywhere else (so far).

First I need to know if I can properly replace that part with the tools I have... I'm not so sure.
 
Hmm... I'm not sure about that. The link you provided is about the internal VGA chip mounted on the motherboard of the ThinkPad T440, T440s, X240, X240s laptops.

How did you link that to the Lenovo 57Y4393 external adapter?
 
To etienne51
You are right,the ANX6212 is difficult to find,last time i checked in September there were lots of shops that sold it.

To Strat_84
That Lenovo converter is interesting,it cost much,we need to contact Lenovo for informations
I knew that Lenovo was mounting that chip on notebooks,but i didn't know the converter
 
Yes you're right Etienne, I first thought about an external converter because it was labeled as "displayport to VGA converter" in the notes, but it's not. It's a chip on a board.

There's no link between them, except that both are built by Lenovo and they may have used a similar chip in the external converter. But that's a big "may"
 
Allright, I see. I'll try to contact them, see if they can tell me which chipset they use in their adapters. But given the specs, it seems like it's limited to 1920x1200 so I'm not sure it's the ANX6212.
 
Well,it seems that someone is interested to make an adapter with the Analogix chipset,i will inform you when i will have more informations.
Analogix is only a chipset maker,is the chinese manufacturer of the adapter that fucked us.
 
Back
Top