Hi,
I was given a Sharp XV-310P projector, and as always nothing is really free as this projector powers up for a second then dies. (by this I mean the fan spins and the 3 LEDs light up quickly).
When I disconnect the HV ballast board for the lamp the unit powers up as it should, so I download the service manual and start testing the hard worked components on the ballast board.
This unit has 4x 2SC4297 NPN triple diffused planar transistors on this board for switching, the problem is they measure ok until I do a reverse E-C leakage test.
When I hook my test bulb which is a 30v PSU is series with a light bulb and do a reverse E-C test the lamp lights, but very dim compared to when I do a forward C-E and apply a 6R resistor to the base to vcc.
Will this matter? The datasheets I found do NOT have anything written about a reverse diode internally and as an actual fact the schematic shows an EXTERNAL diode on the PCB across C-E.
Ok so I am stuck, do these transistors have an internal reverse diode? Are ALL FOUR faulty and leaking? Even though they are leaking reverse E-C will this really matter as there is an external diode to do this? Or do you think the transistor may break down forward C-E if I apply 100v or more?
I appreciate any help on this matter from someone who has delt with beasts like these before.
I was given a Sharp XV-310P projector, and as always nothing is really free as this projector powers up for a second then dies. (by this I mean the fan spins and the 3 LEDs light up quickly).
When I disconnect the HV ballast board for the lamp the unit powers up as it should, so I download the service manual and start testing the hard worked components on the ballast board.
This unit has 4x 2SC4297 NPN triple diffused planar transistors on this board for switching, the problem is they measure ok until I do a reverse E-C leakage test.
When I hook my test bulb which is a 30v PSU is series with a light bulb and do a reverse E-C test the lamp lights, but very dim compared to when I do a forward C-E and apply a 6R resistor to the base to vcc.
Will this matter? The datasheets I found do NOT have anything written about a reverse diode internally and as an actual fact the schematic shows an EXTERNAL diode on the PCB across C-E.
Ok so I am stuck, do these transistors have an internal reverse diode? Are ALL FOUR faulty and leaking? Even though they are leaking reverse E-C will this really matter as there is an external diode to do this? Or do you think the transistor may break down forward C-E if I apply 100v or more?
I appreciate any help on this matter from someone who has delt with beasts like these before.
Hi,
Never worked on the Sharp but transistors do read C-E when reverse biased. Even an AVO 8 on high ohms will do this with any transistor, the 15 volt battery is high enough to break the junction down. Everyone is used to digital meters these days which will not in general do this.
If you pass a high enough current this way you risk damaging the devices. If the device had an internal diode it would read as such and your bulb would light at full brightness. A power transistor will hopefully survive what you are doing to it, small signal ones won't.
Regards Karl
Never worked on the Sharp but transistors do read C-E when reverse biased. Even an AVO 8 on high ohms will do this with any transistor, the 15 volt battery is high enough to break the junction down. Everyone is used to digital meters these days which will not in general do this.
If you pass a high enough current this way you risk damaging the devices. If the device had an internal diode it would read as such and your bulb would light at full brightness. A power transistor will hopefully survive what you are doing to it, small signal ones won't.
Regards Karl
Thank you for the quick reply - Sometimes we forget these simple things! its been a long Friday!
This lamp ballast board only has two two pin connectors going back to the SMPS power board.
One set is coming straight after the bridge rectifier from the 240v mains so shares the SMPS boards filter and bridge, Logic suggests this is not causing the shutdown as no fuses are blown and nothing is monitoring this line and its not drooping.
The other two pins come from an optocouplers phototransistor labeled "lamp off signal" on the service manual, I am not going to try power the ballast board without this optocoupler connection as it is there for protection but I may try bring up a dc voltage from 30 - 200 VDC on a variac to the ballast input with a bridge and filter cap and do some tracing.
The bulb may well be on its way, its a metal halide 80v/120w and will cost AU$700 to replace so I am testing everything else first, I will NOT buy another lamp so worst case is this ends up as parts, I just want to be certain the bulb isn't blown.
Does anyone have any suggestions on how to test such a bulb?? I am taking every step slowly as I repair amplifiers for a living (including SMPS based DVD combo rubbish units) and have had training on safety and logical methods of deduction, but projectors are a new game for me!
Can I just get something close to 80VDC and test the bulb? I am not familiar with metal halide lamps.
Once again thankyou for any suggestions.
This lamp ballast board only has two two pin connectors going back to the SMPS power board.
One set is coming straight after the bridge rectifier from the 240v mains so shares the SMPS boards filter and bridge, Logic suggests this is not causing the shutdown as no fuses are blown and nothing is monitoring this line and its not drooping.
The other two pins come from an optocouplers phototransistor labeled "lamp off signal" on the service manual, I am not going to try power the ballast board without this optocoupler connection as it is there for protection but I may try bring up a dc voltage from 30 - 200 VDC on a variac to the ballast input with a bridge and filter cap and do some tracing.
The bulb may well be on its way, its a metal halide 80v/120w and will cost AU$700 to replace so I am testing everything else first, I will NOT buy another lamp so worst case is this ends up as parts, I just want to be certain the bulb isn't blown.
Does anyone have any suggestions on how to test such a bulb?? I am taking every step slowly as I repair amplifiers for a living (including SMPS based DVD combo rubbish units) and have had training on safety and logical methods of deduction, but projectors are a new game for me!
Can I just get something close to 80VDC and test the bulb? I am not familiar with metal halide lamps.
Once again thankyou for any suggestions.
I have quickly uploaded the relevant schematic if anyone has time to look (I know I know.....)
I am now waiting to see if anyone can give me a hint on testing metal halide bulbs or if anyone has any tips - In the mean time I will desolder some components and do cold testing.
The lamp connects to the bottom right, the rectified mains goes to the top right and feedback is top left.
I am now waiting to see if anyone can give me a hint on testing metal halide bulbs or if anyone has any tips - In the mean time I will desolder some components and do cold testing.
An externally hosted image should be here but it was not working when we last tested it.
The lamp connects to the bottom right, the rectified mains goes to the top right and feedback is top left.
Quick update.
Reading the users manual (sometimes I get so involved with service manuals I forgot that a little bit of research of how the unit operates doesn't hurt!) I found that the bulb is NOT blown.
The user manual states a red error light would come on in this case, not a complete shutdown of the system!
The unit works without this ballast board connected, so this "lamp off signal" is causing this...
The lamp off signal goes back to the main SMPS board, The main SMPS board is taking this signal correctly as when the optocouplers transistor turns on it shuts down, otherwise when I disconnect the board the unit starts up.
Reading the users manual (sometimes I get so involved with service manuals I forgot that a little bit of research of how the unit operates doesn't hurt!) I found that the bulb is NOT blown.
The user manual states a red error light would come on in this case, not a complete shutdown of the system!
The unit works without this ballast board connected, so this "lamp off signal" is causing this...
The lamp off signal goes back to the main SMPS board, The main SMPS board is taking this signal correctly as when the optocouplers transistor turns on it shuts down, otherwise when I disconnect the board the unit starts up.
Thank you Mooly you have been more than helpful already 
.
I am now looking into a few flow charts and I found a schematic with voltages, which doesn't help yet as the PSU shuts down when the ballast board is connected.
Im looking more into the area where the feedback is sent at the moment.
But you have already been helpful so I thank you for that.
.I am now looking into a few flow charts and I found a schematic with voltages, which doesn't help yet as the PSU shuts down when the ballast board is connected.
Im looking more into the area where the feedback is sent at the moment.
But you have already been helpful so I thank you for that.
As far as I can tell at this stage the only way this ballast board can shutdown the SMPS board is by turning on that optocoupler - the only other connection comes straight from 230v after a bridge and cap.
My Friend just scanned the ballast board on a high res scanner and looking up close there are some obvious dry joints!
I have uploaded a picture for anyone interested in another day of my life.
My Friend just scanned the ballast board on a high res scanner and looking up close there are some obvious dry joints!
I have uploaded a picture for anyone interested in another day of my life.
An externally hosted image should be here but it was not working when we last tested it.
A few crusty ones for sure. The large I/C, can't make out on the image, looks like it monitors various rails etc, the usual PSU in a chip job. We all know with anything like this that it's never the I/C but it seems as good a place as any to start (not replacing it ! yet) but checking all its inputs etc.
Have fun
Have fun
me and my mate are just having a quick break, we have touched up all the joints that looked suspect and getting ready to switch on the unit.
By using a Variac with a rectifier into the ballast board we can power the unit up and test voltages without the factory SMPS board shutting everything down.
Oh and for those following this, This ballast board is connected to the SMPS board with two connections, one connection is VIA RELAY CONTACTS to the rectified and filtered mains 230 volts.
One is marked "Lamp off signal" and is the transistor side of an optocoupler, probably signalling a fault causing the main SMPS to shutdown.
Now we will compare voltages if nothing blows up without the opto feedback
all a bit of fun - I know it could have been done differently but those lenses look nice
By using a Variac with a rectifier into the ballast board we can power the unit up and test voltages without the factory SMPS board shutting everything down.
Oh and for those following this, This ballast board is connected to the SMPS board with two connections, one connection is VIA RELAY CONTACTS to the rectified and filtered mains 230 volts.
One is marked "Lamp off signal" and is the transistor side of an optocoupler, probably signalling a fault causing the main SMPS to shutdown.
Now we will compare voltages if nothing blows up without the opto feedback
all a bit of fun - I know it could have been done differently but those lenses look nice
Ok here is what is happening in detail, I disconnect the opto output from ballast board but the SMPS still dies as soon as the ballast board is told to activate, just like before... So I was wrong about the opto signalling a shutdown.
Here is what happens.
-The main SMPS turns on.
-A relay turns on once the voltage is stable.
-The relay contacts turn on the ballast board as shown above.
I think the SMPS suffers a brownout, remember both main SMPS and ballast share the same bridge and cap from the 240 mains.
I think the brownout prot kicks in and LOCKS on to save the bulb - I have to switch off for 5 mins before it will start again.
I will check the rectified mains rail to see how low its dropping on ballast startup and check the resistors used to measure the mains for brownouts on the SMPS IC.
I may connect a 240v 100w in series with the ballast to prove this.
But for now I sleep.
Here is what happens.
-The main SMPS turns on.
-A relay turns on once the voltage is stable.
-The relay contacts turn on the ballast board as shown above.
I think the SMPS suffers a brownout, remember both main SMPS and ballast share the same bridge and cap from the 240 mains.
I think the brownout prot kicks in and LOCKS on to save the bulb - I have to switch off for 5 mins before it will start again.
I will check the rectified mains rail to see how low its dropping on ballast startup and check the resistors used to measure the mains for brownouts on the SMPS IC.
I may connect a 240v 100w in series with the ballast to prove this.
But for now I sleep.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.