AVS Forum banner

Status
Not open for further replies.
1 - 20 of 24 Posts

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #1 ·
So on the advice of another forum member (I won't name any names MIKE! :D) who was able to reduce the electronic noise level a bit on his BG800, I cleaned the pins and resoldered all the components (there aren't many) on the Power Input Module. At the same time I figured it would be a good idea to do the same on the SMPS: Cleaned & resoldered some of the larger components, and cleaned the contact pins with RatShack contact cleaner. (I had already done the transformers a couple of years back and most of caps were replaced 2 months ago so there wasn't much to resolder).


Waited for the alcohol based cleaner to evaporate, plugged it all back in, turned it on, and the SMPS starts to 'sqeal' up but then shuts down after 3/4 of second and then starts over. It does this exactly 11 times then drops back into Standby mode.


While it's doing this, all the voltage diagnostic lights on the Focus Module all light up and fade out with each attempt. So it looks like all rails are supplying their DC....


Figured it was a short somewhere but I've completely checked all traces very closely over the last 2 days. Now I'm starting to think it may have just been bad luck that something blew at that moment...


I pull out the service manual and go through the SMPS power up steps:


(1) Standby voltage lines (+SB, -SB) are ok. (They measure ~8.8V. Supposed to be 9V. Close enough.)


(2) IC102 (TDA4601) pin nine reaches at least the 11V required which turns on the device. Ok.


(3) I see the proper ref. voltage on IC102 pn 4 of around ~4.2VDC which seems fine. Ok. (It's supposed to be 4.4, but my DMM has trouble keeping up the fast rise/fall so who knows).


Things I tried:


- Checked all fuses in the EHT module (1), Power Input Module (2), and the SMPS (9 I think). All ok.

- Tried a new EHT module - no difference. EHT error light doesn't light up either so I guess this is good.

- Tried a new horiz defl module, and a new vert. defl module. Problem still exists.

- Desoldered Q100/Q101 (BUP101) switching transistors and tested them. They're ok. (No opens where there shouldn't be, and no shorts).

- I already replaced 80% of the electrolytic caps about 2 months ago with 105'C rated ones of the same value as some of the ESR's were way out. Remeasured all ESR's again just to be safe. All good.



So what else can I try here? What about IC100 (the other TDA4061) - can I test anything similar there? The service manual doesn't talk at all about IC100.


I really want to avoid starting to desolder and test every single component. Any hints on good places to start? Paths to take? The power off after 3/4 of a second has to be triggered somewhere... is there anything I can measure to determine exactly what the reason for failure is so that I can pinpoint?


A couple of weeks ago the old and wise grandpa Curt had this sage advice to give:

"Don't get me wrong, I'm not raining on your parade, and I'm all for experimenting (see our local waste disposal area for the results of many of my 'experients'..;-), but sometimes even logical component replacement makes things worse than using the parts of the original manufacturer."


I should have listened to him! :)



Kal
 

·
Registered
Joined
·
4,018 Posts
I'm no Barco expert, I just play one on TV.

See if there is a small capacitor on the PS, between 1mfd and 4,7 or so. Often this is used as a Kick-start. Check there. If that's ok, you may have a short somewhere as you started to diagnose. Ya shouldn't have touched it.


Marc
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #3 ·
Quote:
Originally posted by Tinman
I'm no Barco expert, I just play one on TV.

See if there is a small capacitor on the PS, between 1mfd and 4,7 or so. Often this is used as a Kick-start. Check there. If that's ok, you may have a short somewhere as you started to diagnose. Ya shouldn't have touched it.


Marc
Thanks Marc! But all of the DC rails start up just fine (the DC rail test lights on the Focus Module light up) so I don't think it's a kick start problem (?). Problem is that after 3/4 of a second they all shut down and start over...

Or maybe I mistunderstood what you meant (?).


I forgot to mention before that the startup squeal also sounds different now (for that 3/4 of a second). Before I had the familiar sound that was higher pitched right from the start. Now for the 3/4 of a second it sounds more like a sqawk or a kazoo instead of a squeal.


Kal
 

·
Registered
Joined
·
1,745 Posts
Was the SMPS working ok before you messed with the power input module? Did you say you resoldered components on the SMPS when the power input module was being worked on? You should make absolutely certain the solder joints on the SMPS are good now. Your description makes it sound like the oscillator that drives the big SMPS power transformer is not running the correct frequency (different kind of "sqwak"). My best guess is this could happen if one of those SMPS capacitors got accidentally desoldered out of its tuned circuit.


I doubt you have a short because a short would cause a fuse to blow or smoke to rise and you haven't seen either. Seems like your oscillator is stalling out because is it running the wrong frequency.


Just my opinion I am no expert either.
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #5 ·
Thanks for the input Bob! Some good points you make there on the 'squawking' vs. 'squealing'.


Both items worked fine before. The projector was used for 4-6 hours Saturday night, then Sunday morning I pulled out the Power Input Module and the SMPS and resoldered connections and cleaned the contacts on both. Plugged them both back in and that's when the problems started.


Kal
 

·
Registered
Joined
·
18,503 Posts
Check all switching diodes, all voltage rails for shorts. The 1000uF caps @ 40V on the motherboard right under the main PS sometimes short out for no reason.


Gramps out!


Curt
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #7 ·
Thanks for the hints Curt! I had gone through a month or so ago and checked the ESR on every single 'lytic cap in the PJ and ended up replacing about 50 of them (105'C, same uF rating, one step up in voltage).


Just to be sure I checked those 3 1000uf'ers on the backplane - no leaks, no shorts, ESR still reasonable. The one closest to the front did measure a bit leaky (didn't seem to charge up right as the resistance across it never went over 0.5K) until I took it out of circuit (what a b!tch when you're ceiling mounted!)... it must be in series with a 0.5K resistor somewhere...


No problems with the diodes either: I've gone through all of them in the SMPS but didn't measure any out of circuit (just made sure that I measured around .4-.6V drop across with my DMM's diode setting in one direction and considerably more the other way). A couple measured odd so I tried them out of circuit but they were fine.


I've tried different EHT/horiz sync/vert sync/horiz shift cards to make sure the problem wasn't there.


Just to be 100% sure it's the SMPS, can I safely run the SMPS with some of the other cards pulled out? If I unplug the CRT's from their sockets and unplug the focus module, quadrupler, HV splitter, RBG input, Quad decoder, and RGB outputs will the SMPS still try to run?


Kal
 

·
Registered
Joined
·
968 Posts
Kal,


I know I've run without the 800's TTL input card and the convergence board periodically. In fact, I've seen convergence boards that kept the unit from powering up as well.


The other board I regularly forget to reinstall is the EHT board. With this one out you have no high voltage at all, but the tube heaters are still on along with everything else. I just verified this works on a BG800 too.


I really don't have the experience to say what's safe about having HV on, but with tubes disconnected from the splitter. I generally never disconnect more than two at any given time and I tape over those empty deep holes. 34 kVdc is of course coming in on the 4th lead (never disconnect) and I tend to want it to have a place to go other than into three open high impedance circuits so I always leave one tube connected as a load. Like I say, I'm not sure if its required...


If you read the service manual for the quad decoder you'll see that it mentions that it provides spot burn protection. Something to do with the 17 Vdc rail falling off faster than the 230 Vdc rail that drives your video input signal. You might want to keep it in, but if you have no HV running in the first place maybe its okay?


Also, I believe I recall an experience where I set the horizontal width voltage too high and it kept the SMPS from running up. I set it back down a ways and then readjusted it once it was on... but I had to cycle power to get it to try to come on each time.


I'm currently tinkering with two BG800's so if I think of something else I'll let you know.


Mark
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #9 ·
Thanks Mark - I did hear about how you can run the units without the TTL & convergence boards - in fact, I haven't had the TTL board in my unit for about 4 months now. Just another card that nobody uses that draw extra power from the SMPS so why leave it in? Leaving the convergence board out is a good way to test if that's indeed the culprit.


I think I won't risk turning it on with out cards out if I can't confirm that they're not 100% needed. I don't need to blow something else while I'm trying to confirm that my SMPS is defective! :)

"Also, I believe I recall an experience where I set the horizontal width voltage too high and it kept the SMPS from running up. I set it back down a ways and then readjusted it once it was on... but I had to cycle power to get it to try to come on each time."


Do you remember if the power was cycling up for about 3/4 of a second, then shutting down, and then trying to restart again & again? You never know, I may have inadvertently hit the pot by mistake that controls the +17V signal (horiz width)... good hint!


Kal
 

·
Registered
Joined
·
968 Posts
Kal,


Any luck yet?


I was able to recreate the max horiz voltage thing today. It just goes into scan fail but does come on.


After re-reading parts of your descriptions I'm leaning towards a short/bad connection of some sort or the start-up of IC102 as well. R121 could be seeing too much current in Q101 and shuts down IC102 and then it tries again? There are probably plenty of places to draw too much current and not blow a fuse.


Do you have a storage oscilloscope? Or just the DMM?


As far as cards to run without, the EHT is one I'm comfortable with.


However, just for the fun of being uncomfortable yesterday I pulled almost everything except the SMPS and a few other items. It ran, but was also a bit uncomfortable. I sort of assume its because of the half of the SMPS that is frequency dependent and without the horiz deflec board it may be lost.


Have you checked C1? Or from the GNDM side to the other ground? If there were a breakdown from the floating GNDM (-150 Vdc with respect to chassis) and the regular grounds on the output of the SMPS it would not work, but I don't know if it would act like what you're seeing.


Mark
 

·
Registered
Joined
·
4,018 Posts
Kal, have you thought of checking the caps again? Just because they are new does not mean that on can't have shorted. Seen it...



Marc
 

·
Registered
Joined
·
4,018 Posts
You watch... it is probably going to be something real simple you will kick yourself over.


Marc
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #14 ·
Hope you're right Marc!


I've only had an hour to go through so far and I've started with the daughterboard - desoldering the items one at a time and testing them.


About 1/3 of the way through and found one oddity but not sure if it would actually have an effect:


1/2 watt resistor R48 is listed as 27Meg, but measures 37Meg out of circuit (on a DMM that reads up to 2000Meg). It's used as the top end of a voltage divider with R25 (120K) to feed the HTHD voltage into one of the two inputs of a comparitor (IC3).


The difference means that 1/308th of the HTHD is feed in instead of 1/225th as it should be....


I'll keep testing and make a list of replacement parts to buy, though some of these more complex components (IC's, variable zeners, etc.) I really don't know how to beging to test....


Well, at least I'm learning something! :)


Kal
 

·
Registered
Joined
·
18,503 Posts
You can always send it to me for a quick turnaround if you get stuck.


The difference in the 27/37 megohms won't do it.


Curt
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #16 ·
Thanks Curt! Good to know... I may just take you up on that offer! I'm going to spend whatever free time I have tomorrow doing the best I can on this... then I may call you Monday if this SMPS is still giving me headaches.


I'm not all that familiar with the loading that the other cards have the SMPS - I've changed the following list of cards (I had borrowed spares from Mikedd) and it didn't help:


- EHT

- Horizontal Deflection

- Sync + Vertical Deflection

- N/S Correction


Does this pretty much guaranty that the problem's in the SMPS or could it be something else I've overlooked?


Kal
 

·
Registered
Joined
·
7,136 Posts
Discussion Starter · #17 ·
WOOHOO! Curt's da man!!!


Well there you go! It took Curt's eagle-eye to find the problem! It was a solder bridge between two of the GND lines on two of the fuses under the daughterboard.


For the love of god I looked at the traces/solder points so closely (many hours) but never saw anything....! (At my own defence, Curt tells me that it did take him quite some time to find as well) :D


Thanks Curt!


I was really starting to doubt that the problem was indeed in the SMPS....


Kal
 

·
Registered
Joined
·
4,018 Posts
I repeat my above statement :)


Congrats.


Marc
 

·
Registered
Joined
·
4,018 Posts
I have done little things like that often enough that I picked up a stereo microscope on ebay. It has made it much easier to find those darn cracks and solder bridges. I really LOVE those self-induced failures. :)


Marc
 
1 - 20 of 24 Posts
Status
Not open for further replies.
Top