Maker Pro
Maker Pro

Zenith DLP Set Not Working After Lamp Replacement

C

Chris F.

Jan 1, 1970
0
I posted a message about this a few weeks ago, regarding replacement of
the blown lamp. After installing the new lamp, the set still won't start up;
the "replace lamp" error light flashes instead. Checking the power input to
the H.V. power supply (or lamp ballast I suppose), I measured a peak of over
300 VDC initially but it quickly drops to (and stays at) about 160. The 450
volt electrolytic cap connected across the power input suggests that the
operating voltage should be much higher than 160.
So what now? Is this a ballast problem or something on the main power
supply? I can't afford to gamble any more money on this thing.
The model and chassis numbers BTW are E44W46LCD and MB-042B respectively.
 
F

Franc Zabkar

Jan 1, 1970
0
I posted a message about this a few weeks ago, regarding replacement of
the blown lamp. After installing the new lamp, the set still won't start up;
the "replace lamp" error light flashes instead. Checking the power input to
the H.V. power supply (or lamp ballast I suppose), I measured a peak of over
300 VDC initially but it quickly drops to (and stays at) about 160. The 450
volt electrolytic cap connected across the power input suggests that the
operating voltage should be much higher than 160.
So what now? Is this a ballast problem or something on the main power
supply? I can't afford to gamble any more money on this thing.
The model and chassis numbers BTW are E44W46LCD and MB-042B respectively.

FWIW, I have a service manual for an LG MB-042C chassis.

The DC supply voltage for the ballast is 370 min, 390 typ, and 410
max. This is derived from directly across the main 220uF/450V bulk
capacitor (C820). It is not regulated.

The parts list includes ...

6913V00002C BALLAST, EUC100 P/11 PHILIPS 100W POWERPACK BALLAST FOR
LCD

6912V00006C LAMP, HIGH PRESSURE MECURY UHP 100W 1.0 PH P22 PHILIPS
100V 1A RN48SZ40H

- Franc Zabkar
 
C

Chris F.

Jan 1, 1970
0
So based on that info, I should suspect the power supply isn't supplying
enough voltage to the ballast?
 
F

Franc Zabkar

Jan 1, 1970
0
So based on that info, I should suspect the power supply isn't supplying
enough voltage to the ballast?

I don't know. If the ballast is drawing too much current, then I
suppose it's possible for the DC supply to sag significantly. BTW,
when I wrote that the 400V supply is not regulated, I should have said
that this supply is derived from the rectified mains and passes
through an active PFC circuit.

Maybe you could insert a 1 ohm 10W series resistor into the DC feed to
your ballast and measure the voltage drop across it. That would tell
you the current draw.

- Franc Zabkar
 
C

Chris F.

Jan 1, 1970
0
What I did so far was just disconnect the lamp ballast. The voltage across
C820 is still only 160V, and I also determined that all voltages derived
from the switching supply (16, 32, etc) are proportionally low - the 16V
supply is only measuring 6V, for example.
I don't know too much about PFC circuits but I assume something has to
take place to get the correct 400V across that cap, something which clearly
isn't happening at this point.
If you could Email me the schematic of that power supply I'd really
appreciate it. I've got almost $150 in this thing now and giving up just
isn't an option right now.
 
F

Franc Zabkar

Jan 1, 1970
0
What I did so far was just disconnect the lamp ballast. The voltage across
C820 is still only 160V, and I also determined that all voltages derived
from the switching supply (16, 32, etc) are proportionally low - the 16V
supply is only measuring 6V, for example.
I don't know too much about PFC circuits but I assume something has to
take place to get the correct 400V across that cap, something which clearly
isn't happening at this point.

I also have no experience in this area, but I wonder if a leaky D811
could be causing C820 to be discharged by Q801 during PFC cycles ???
If you could Email me the schematic of that power supply I'd really
appreciate it. I've got almost $150 in this thing now and giving up just
isn't an option right now.

This is the PSU section:
http://www.users.on.net/~fzabkar/MB-042C/

The whole manual is about 24MB zipped. I have uploaded it to
eserviceinfo.

To get all 13 parts, replace 370nn with 37046-37052 and 37054-37059 in
either of the following URLs:

http://www.eserviceinfo.com/download.php?fileid=370nn
http://www.eserviceinfo.com/downloadsm/370nn/LG_RT44SZ80LB.html

- Franc Zabkar
 
F

Franc Zabkar

Jan 1, 1970
0
I also have no experience in this area, but I wonder if a leaky D811
could be causing C820 to be discharged by Q801 during PFC cycles ???

I'm assuming you have checked that C820 has not gone open or low in
value ...

- Franc Zabkar
 
F

Franc Zabkar

Jan 1, 1970
0
C820 is good, haven't checked D811 yet. From what I figured out so far,
IC810 has to serve two purposes; first, to drive the PFC circuit which
creates the 400VDC, and second, to drive/regulate the switching power supply
which supplies the low-voltage areas. The gate of Q830 has a strong 1V p-p
signal, but there's nothing at all on the gate of Q801, except of course for
the very short burst when the set is first plugged in. What I'm not sure of
is whether or not these two sections of the IC are dependant on each other;
if not, I could just remove the PFC section and insert 400VDC onto C820 from
an external power supply, just for testing purposes of course.
Is it possible that IC810 is defective? I can't seem to come to any other
conclusion right now.

I don't have any experience with PFC circuits, but the first of the
following two datasheets has an application circuit that describes a
two-chip design:

http://focus.ti.com/lit/ds/symlink/ucc28600.pdf
http://focus.ti.com/lit/ds/symlink/ucc28051.pdf

One chip does the active PFC stuff while the other is a quasi-resonant
PWM controller.

The SMA-E1017 is described as a "PFC&QRC COMBO". I haven't been able
to find its datasheet, but the "ST" in the parts list would suggest
that it is made by ST Microelectronics.

I would think that Q830 would be getting maximum gate drive because
IC810 is sensing a low output on the secondary, so that part of the
chip appears to be working. If Q801 is not being driven, then I would
think that the circuit would be behaving like an ordinary non-PFC
circuit, in which case it would make no difference if you open
circuited the drain.

Have you measured the voltage across C805? You say that you are seeing
160V across C820. My calculations suggest that, if C820 were OC, then
you should measure an average DC voltage of ...

2/pi * sqrt(2) * 240V = 216V

So either something doesn't make sense, or I'm misunderstanding how a
multimeter works.

BTW, are you posting from Canada, as your IP suggests, in which case
how are you expecting to get 400VDC from a 120VAC supply ... assuming
your circuit is similar to the one I have posted? I'm posting from
Australia where the mains is 240VAC, BTW. Does your PSU have a voltage
doubler section at or near the bridge rectifier (D801)??? Is there a
connection between the neutral conductor and the midpoint of two bulk
caps??? If there is no voltage doubler, can the E1017 IC boost the
supply from 160V to 400V via T801??? If so, then you may want to check
R810 and R811 for OC. BTW, the notes around T830 indicate that there
are North American and "domestic" versions of this SM transformer, so
this would suggest that the PSU circuit applies universally. In fact
the Australian manual states that the acceptable input voltage range
is 110VAC-240VAC (10%) at 50/60 Hz. So maybe IC810 performs PFC *and*
voltage doubling???

If it helps, some searching would suggest that your set may be a
rebadged LG 44SZ81 or RU44SZ80L. The manual I uploaded was for an
RT44SZ80LB.

- Franc Zabkar
 
C

Chris F.

Jan 1, 1970
0
C820 is good, haven't checked D811 yet. From what I figured out so far,
IC810 has to serve two purposes; first, to drive the PFC circuit which
creates the 400VDC, and second, to drive/regulate the switching power supply
which supplies the low-voltage areas. The gate of Q830 has a strong 1V p-p
signal, but there's nothing at all on the gate of Q801, except of course for
the very short burst when the set is first plugged in. What I'm not sure of
is whether or not these two sections of the IC are dependant on each other;
if not, I could just remove the PFC section and insert 400VDC onto C820 from
an external power supply, just for testing purposes of course.
Is it possible that IC810 is defective? I can't seem to come to any other
conclusion right now.
 
F

Franc Zabkar

Jan 1, 1970
0
In my case, it does appear that IC810 also performs some kind of
voltage-doubling, via T801. There are no other components to do the job, in
fact the circuit appears to perfectly match the schematic you provided me.
I've since tried injecting 400V from an external power supply (to C820),
surprisingly it doesn't make any difference at all. While the ballast now
gets the correct 400VDC, the output voltages in the low voltage supply are
still down by more than 50%. What's more, the "replace lamp" light is still
blinking. I'm more confused than ever now.
I'm only using that basic schematic you sent me first - I haven't been
able to download the rest because eserviceinfo is not responding today.
Maybe I need a break from this anyway.
I guess a tech should never assume a blown lamp is just a blown lamp, when
working on junk like this.....

Would an overload on the secondary side cause IC810 to shut off its
PFC and voltage doubling function? Maybe you should try unloading some
of the secondaries, eg the sound circuit.

Otherwise you may want to check what IC830 and IC851 are doing. I
would expect that, because the secondary outputs are low, this should
cause the optocoupler's diode to be driven hard, which should in turn
mean that pin 4 should be close to ground. This should at least prove
whether the voltage feedback is working. I would then check R837. This
resistor determines the current limit of your circuit. IC810 could be
reducing the drive to Q830 in response to an overload on the
secondaries which it would sense via R837.

- Franc Zabkar
 
F

Franc Zabkar

Jan 1, 1970
0
Otherwise you may want to check what IC830 and IC851 are doing. I
would expect that, because the secondary outputs are low, this should
cause the optocoupler's diode to be driven hard, which should in turn
mean that pin 4 should be close to ground.

Oops, it should be the other way around. The diode will *not* be
driven, in which case pin 4 should be high.

- Franc Zabkar
 
F

Franc Zabkar

Jan 1, 1970
0
In my case, it does appear that IC810 also performs some kind of
voltage-doubling, via T801. There are no other components to do the job, in
fact the circuit appears to perfectly match the schematic you provided me.
I've since tried injecting 400V from an external power supply (to C820),
surprisingly it doesn't make any difference at all. While the ballast now
gets the correct 400VDC, the output voltages in the low voltage supply are
still down by more than 50%. What's more, the "replace lamp" light is still
blinking. I'm more confused than ever now.
I'm only using that basic schematic you sent me first - I haven't been
able to download the rest because eserviceinfo is not responding today.
Maybe I need a break from this anyway.
I guess a tech should never assume a blown lamp is just a blown lamp, when
working on junk like this.....

Doh, I just realised that your set is behaving as if it is in standby
mode. Notice the "POWER ON" signal at pin 1 and the ST-5.5V rail at
pin 2 of PB41 which goes to the Tuner Board. I suspect that if you
disconnect PB41 and short pin 1 & 2, the PSU will power up.

I would take care when doing this because disconnecting PB41 also
removes power to the fan.

- Franc Zabkar
 
C

Chris F.

Jan 1, 1970
0
BTW, are you posting from Canada, as your IP suggests, in which case
how are you expecting to get 400VDC from a 120VAC supply ... assuming
your circuit is similar to the one I have posted? I'm posting from
Australia where the mains is 240VAC, BTW. Does your PSU have a voltage
doubler section at or near the bridge rectifier (D801)??? Is there a
connection between the neutral conductor and the midpoint of two bulk
caps??? If there is no voltage doubler, can the E1017 IC boost the
supply from 160V to 400V via T801??? If so, then you may want to check
R810 and R811 for OC. BTW, the notes around T830 indicate that there
are North American and "domestic" versions of this SM transformer, so
this would suggest that the PSU circuit applies universally. In fact
the Australian manual states that the acceptable input voltage range
is 110VAC-240VAC (10%) at 50/60 Hz. So maybe IC810 performs PFC *and*
voltage doubling???

In my case, it does appear that IC810 also performs some kind of
voltage-doubling, via T801. There are no other components to do the job, in
fact the circuit appears to perfectly match the schematic you provided me.
I've since tried injecting 400V from an external power supply (to C820),
surprisingly it doesn't make any difference at all. While the ballast now
gets the correct 400VDC, the output voltages in the low voltage supply are
still down by more than 50%. What's more, the "replace lamp" light is still
blinking. I'm more confused than ever now.
I'm only using that basic schematic you sent me first - I haven't been
able to download the rest because eserviceinfo is not responding today.
Maybe I need a break from this anyway.
I guess a tech should never assume a blown lamp is just a blown lamp, when
working on junk like this.....
 
C

Chris F.

Jan 1, 1970
0
You're right, and I should have realized that a long time ago. Shorting
those pins did power the PSU and all voltages are now normal, but the lamp
still does not come on. I guess the next step is to test the ballast, but
for that I need the pinout, and eserviceinfo is still down.......
 
F

Franc Zabkar

Jan 1, 1970
0
You're right, and I should have realized that a long time ago.

I was thrown by the fact that all the voltages were at 40%. You would
normally expect that all rails would be at 0V apart from a single
standby supply.
Shorting those pins did power the PSU and all voltages are now normal ...

If the ballast supply is now at 400V, then this would imply that the
PFC and voltage doubling function of IC810 is turned off in standby
mode and turned on in run mode. This begs the question, how does IC810
know that the set is in standby mode? Perhaps this is the function of
pin 5?
... but the lamp still does not come on.

The uP on the Tuner Board appears to control and sense the lamp via
connector P10. BTW, when you shorted pins 1&2 of PB41, I hope you
removed the connector, otherwise one of the uP pins would have been
subjected to 16V.
I guess the next step is to test the ballast, but
for that I need the pinout, and eserviceinfo is still down.......

I have uploaded tuner PCB circuit 2-of-2 to my web space:
http://www.users.on.net/~fzabkar/MB-042C/

It seems to me that if you can't get the set to come out of standby,
then the M37151 uP (IC1) may be faulty, in which case your lamp
problem may be a secondary issue, if it is in fact a problem at all.

- Franc Zabkar
 
C

Chris F.

Jan 1, 1970
0
Franc, you're not going to believe what was wrong the whole time.... I
hadn't yet replaced the cover to the lamp compartment, and last night I
realized there was a switch under that cover to sense when it was open.
Replaced the cover and the set works fine! Can't believe I was so dense as
to not realize that right from the start, instead of going on this wild
goose chase. I should have realized that the toughest problems often have
the simplest solutions.
I still thank you for your help though, it's good to know there are folks
out there to help me when I really do need it.
 
F

Franc Zabkar

Jan 1, 1970
0
Franc, you're not going to believe what was wrong the whole time.... I
hadn't yet replaced the cover to the lamp compartment, and last night I
realized there was a switch under that cover to sense when it was open.
Replaced the cover and the set works fine! Can't believe I was so dense as
to not realize that right from the start, instead of going on this wild
goose chase. I should have realized that the toughest problems often have
the simplest solutions.

I suppose one day you'll look back on this job and laugh about it.

I learned a lot in any case, so my time wasn't wasted.

- Franc Zabkar
 
C

Chris F.

Jan 1, 1970
0
I might not be out of the woods just yet.... although the set is working,
the remote isn't responding and I can't get a universal remote to work
either. The service manual does not include the control panel board so I
have no way to trace the remote signal. I'm not sure what to do now, I'm
worried that this may be something I caused while I had the set apart.......
 
F

Franc Zabkar

Jan 1, 1970
0
I might not be out of the woods just yet.... although the set is working,
the remote isn't responding and I can't get a universal remote to work
either. The service manual does not include the control panel board so I
have no way to trace the remote signal. I'm not sure what to do now, I'm
worried that this may be something I caused while I had the set apart.......

Pins 1,2,3 of P2 are where the IR sensor connects to the tuner board.

See http://www.users.on.net/~fzabkar/MB-042C/Tuner_UL.jpg

If you've disconnected P2, then you'll also have disconnected one of
the I2C buses.

- Franc Zabkar
 
F

Franc Zabkar

Jan 1, 1970
0
I was going to remind him of this myself until he told us that his
remote doesn't work.

I have a Powerpoint presentation that describes how to reset the
counter and how to recognise the difference between 5kV and 15kV lamps
and ballasts. I can upload it if needed.

- Franc Zabkar
 
Top