CurtPalme.com Home Theater Forum

**Curt Palme** · Posted: Tue Nov 10, 2009 1:58 am Post subject: Obscure Ampro question for tse.

Here's one I haven't run across before.

I have that low hour 4600 that I have test running on the bench, and got in a CPU and HV splitter from another 4600 that I wanted to test.

The HV splitter is fine, no problems there, but the customer's CPU board gives a -25 volt error message. He complained that he got a +25 volt high message in his set, so he wanted me to check it.

I have not yet measured the 25 volt line, but am curious if the CPU board itself can develop an error message when there really isn't a problem?

The original CPU board works fine, no error messages.

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

Ouuuu...I know I know....oh wait.....I'm not TSE.......never mind...... Sad

**Curt Palme** · Posted: Tue Nov 10, 2009 2:14 am Post subject:

Oh c'mon, spill it! Smile

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

Ok.... since you asked... Laughing

The direct answer is yes any error can be triggered by the CPU board even if the problem doesn't exist. All voltages are sampled by the CPU board so any break in that sample circuit can trigger the error code.

For those 2 codes I would start simple and check the board socket pins to make sure no pins are bent over and solder joints look good. Then I would see if the 5 volt led is lit and solid. The voltage samples are referenced off the CPU's onboard 5 volt supply. Next I would ask you if IC1 is socket-ed on that board. Some were and some weren't.

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

Alright I just check my sheets and IC1 should be a PLCC chip so I would pull it and clean chip and socket but I would also examine the socket contacts. I've had some of these CPU boards where the contacts were pushed in and not making contact. For those codes make sure you check pins 18 and 79.

Robert A. Hill · Joined: 19 May 2006 Posts: 182 Location: Simpsonville, SC

Mac, spill the beans! Were you an Ampro service tech? I thought you were an electronics tech for consumer electronics.

MikeEby · Joined: 24 Jun 2007 Posts: 5237 Location: Osceola, Indiana

Robert A. Hill · Joined: 19 May 2006 Posts: 182 Location: Simpsonville, SC

Mike,
I agree with you 100%. Two of my sons are techs and I always tell them to keep their skills polished even in our disposable society. I have trouble finding almost any service manual....without paying for it...and really appreciated Curt's manuals and information. It seems that Mac has every schematic every printed for everything! To me, that is amazing.

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

LOL.... no I never worked for Ampro.

But I've hand my hands on a number of them along with Sonys, NECs, Zeniths and I forget what else. And all kinds of consumer electronics. I dont always have a service manual on hand but I have access to almost all if need be. Give me a print and I can talk you through the repair. No print then give me the board and I'll find the problem. The print just shortens the time locating the problem.

As far as the Ampro CPU boards, a while back I got a bunch of non working ones so I spent some time going over their design and also started saving various firmware and bought a bunch of blank chips. I like the design because you can take a CPU board out of one projector, flash a chip with the firmware of the projector your putting it in and WALL LA. You can even change what remote you want to use. 2300/2600, 3300/4300 and 3600/4600 cpu boards all interchange. In the end all the cpu boards were repaired so I have stock on hand..... Smile

**Curt Palme** · Posted: Tue Nov 10, 2009 7:57 pm Post subject:

OK, FWIW, that didn't work. I reseated and cleaned all socketed chips, and still get the -25 volts high error message on the customer's CPU board.

We're dealing with the set blowing fuses whenever three HV leads are connected, so I'm working on that more than I am the CPU board. I'll tell him to use it as is for now.

Mac's a damn smart cookie, one day I need to go to his neck of the woods to see how he works.

It occurs to me that between Tinman. Marc, John in France, myself and a few others, we could open an electronics repair shop that repairs just about anything. The problem is, it would never make money, so it's not viable, but the brains around this forum never ceased to amaze me.

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

Ok, I'm trying to remember which error you get that goes away when you turn off registration. Let me check my info, I think I have it here somewhere.

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

Yep, thats the ones. Try turning registration on or off and see if the code goes away. Even if not its still ok to run it like that if its not causing any problems.

tse · Joined: 03 May 2006 Posts: 1014 Location: Sweatbucket, Fl.

Pretty good answers. About the only thing that runs from the -25V rail is the registration amp. If the registration is turned off or the registration is set to midrange there is very little load on the -25V coming from the SMPS. The -25V might actually be high in this situation. The registration amps are rated for +/-30V so not really a problem if not too much more than -30V.

This was a learning experience. Some SMPS would have high outputs with registration turned off some were ok. It turned out to be the transformer. After disecting a couple it was found that the transformers with the 25V windings that were wound with the turns really tight together had poor load regulation. Those with sloppy turns that covered a bigger area of the bobbin worked better! The spec was changed to evenly spread the turns across the bobbin and voila! Much better load regulation.

Attached is a pick of the feeble attempt to provide a load when the output wanted to go high. A 24V zener in series with a couple hundred ohm resistor. It kinda worked but some supplies till shipped with the 25V high error when the registration was turned off.

Scott

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

Whats interesting here Scott is that (if the provided info is correct) he's only getting the error with 1 CPU and not the other CPU. The code seems to be CPU dependent. The error triggering must have a slight variation between the 2. Also involved on the sample circuit of the cpu is a resistor pack and diode. Other wise it goes directly to the processor.

Maybe the original problem was more cpu related then SMPS related. I'm curious if it goes away when registration is turned on.

**Curt Palme** · Posted: Wed Nov 11, 2009 1:32 am Post subject:

I packed the set away for now, it ran fine otherwise, and I've got other stuff to do. I will put it back on the bench in the next day or two and will post.

macgyver655 · Joined: 22 Aug 2007 Posts: 8508

I had a hunch so I went and looked up something. Curt, try swapping the firmware chip from the good cpu to the bad cpu and see if the error code goes away.

stefuel · Posted: Wed Nov 11, 2009 11:08 am Post subject:

I've read this whole thread and the problem can't be the SMPS if the CPU is reporting the same error on two different sets.
I would just sell him another CPU and swap his firmware into it and call it a day.