What's new

Hatsune Mike

Champion
Joined
Dec 8, 2017
Messages
736
Reaction score
2,122
Location
Tokyo, Japan
After about five minutes of being powered on with a tube I bought recently, with Twinkle Star Sprite's attract mode, my MS-2930 chassis had the screen suddenly go black (but not collapse as during a normal power-off), with audible static sounds from the tube face (like during a normal power-off!) The neck glow and deflection seemed to continue during this time. No pop sound, smoke, or smells.

I pulled the chassis, gave it a look-over, and cleaned up the black goop (whichwe have known for some time is harmless, just some potting compound from a nearby component).

After the cleanup, I powered it on. The black level was higher, making the image grey, as if SCREEN had been turned up (it's possible I nudged the knob during the cleaning though). The image puncushion was a little off, and the whole image became smaller, about 1" in on each side.

I have not measured yet, but the smaller image and warped geometry makes me think the B+ power supply has suddenly changed. I wonder if the shutdown at first was from X-ray protection acting up.

I have a 2930 parts chassis I can try swapping parts from, but I am not sure where to start. Has anyone seen these issues?
 
I had a few questions about your experience:

-What did you use to clean up the black goop?
-Where was the goop located on the chassis?

-After the cleanup, when your image powered on, did you try adjusting SCREEN, or did you observe the differences and power it off?
 
I used isopropyl alcohol to clean up the goop. It was located near the flyback, but had originated from the grey component on the other side of the heatsink (with the HOT attached).

After cleaning up and powering it on, I adjusted SCREEN to bring the black level down to normal. However, I don't believe I nudged screen, and I definitely didn't nudge the B+ or pincushion pots, so I'm led to believe the off black level and new issues are a byproduct of something else.
 
-Can you calibrate the chassis back to its original state, or does it stop you at certain levels?

-Just to be sure, are you using 99% percent isopropyl alcohol when cleaning?

The reason I ask is because I've killed a few electronics in the field with lower percentages. Doesn't mean that's what happened, but I'd like to rule it out.
 
Regarding the alcohol, I do not think we need to regard the percentage, as I only used it on the PCB itself. To clean it, I removed the flyback from the PCB first. So, I did not put it on electronic packages.

I expect I could calibrate it back to the "correct" size and geometry, but my fear is of causing further damage to the chassis by running it in a compromised state.

I'm looking now to see where I should be measuring B+, and what the correct level should be. I am also turning my attention towards another 2930 I have that "works" but has no RGB. First step is inspecting the regulated 12VDC going to the RGB amp.
 
I recapped the chassis, and it... seems stable? I noticed the HV adjustment knob was turned way up (which I did not do) so I put it back to the factory position, which was easily determined by the silicone glue that was still attached.

Here is a video of it with a Neo-Geo on my junky test tube
:

The tube I am testing with is trash, it was dropped before I ever got it, so the discoloration can be disregarded. I just use this one for testing in case a freak accident should harm a good one (loss of deflection, HV overvoltage, etc)

I think I will cautiously install this in my Blast with my tube, and just keep a close eye on it.
 
So a complete recap and adjustment of the HV fixed it so far. I wonder why someone had it up so high in the first place?

Got a video of it running in the Blast?
 
Captured this this morning:

It looks so good that I’m suspicious. I need to measure the heater voltage and make sure the filament is not being overdriven for a (temporary) contrast boost.
 
The heater voltage comes off the main power transformer on the PCB, and only goes through a resistor. It should be a 60Hz sine wave, so I think any regular multimeter can measure it without issue.
 
I don't off-hand (I will have to get home, and then remove the chassis...) but you can trace the "H" pin from the 4-wire ribbon on the neck board down to a large resistor on the main chassis, not that far from the flyback. Ultimately, it ends up at the large transformer that straddles the primary/secondary supply.
 
Okay, did some sleuthing.

The heater voltage leaves the chassis at CN703, close to the RGB amp at U201. Tracing that back on the underside, it leads back to R714, to the left of the flyback. On the other side of the resistor, it’s connected to EY57, the “out” leg of diode D955. This is in parallel with C956. Together, these two components form a half-wave rectifier, with the capacitor smoothing the voltage.

As this is the design, we can ostensibly measure it as a DC voltage, although there may be some 60Hz ripple since it’s a crude rectification.

With a multimeter, I measure 6VDC across pins 9 and 10 on the neck (the filament), and 0.1VAC. I’m shaky on electrical theory, but I think that AC measurement gives us a very rough estimate of the ripple voltage.

My heater resistor looks stock, as it matches a spare chassis, and the voltage is very close to the ideal 6.3V. I’m no longer worried!
 
ugh, okay, never mind. Still worried about the chassis.

I have pretty consistent repro for the HV shutdown:

If I turn the contrast down a bit, then it doesn’t happen, or at least not until a fully bright scene. Seems related to current from the power supply.

Another fun issue: I turned down the HV slightly and eased off on the HV protect, and now I have this fun issue. Happens consistently, but I cut power immediately for fear of damage.

 
Is it still that spare tube or the original blast one?

Could be a faulty Transistor that regulates the voltage on the chassis.
 
Is it still that spare tube or the original blast one?

Could be a faulty Transistor that regulates the voltage on the chassis.
99% sure that was HV shutdown, and the HV protection was kicking in. It makes sense as to why it'd go larger when you think of it this way, which a friend mentioned:

  1. HV shutdown kicks in. Voltage entering flyback begins to be reduced.
  2. HV voltage (~30kV) starts to decline, and focus voltage starts to decline proportionately
  3. Image grows dimmer and out of focus
  4. As HV voltage declines, attraction of electrons to the front of the tube is reduced, so travel speed is reduced
  5. Reduced travel speed means particles spend more time near the yoke's area of strongest influence
  6. As a result, deflection angle is increased, causing the "zoom" effect seen in the video

I was able to change this behavior by changing the HV adjust knob, rather than B+. Sharp Image still has this chassis "on the bench" (must be a large and crowded bench).

That video was on the tube that came with the MS-2930 originally. There's a lot of FUD around Blast tube matching, yoke types, and heater resistors. In my experience, any MS-293x tube in good condition, with the right horizontal yoke impedance (~0.2mH) has had consistent performance with any Blast chassis. I measured the heater voltage to be ~6.3V consistently each time I've tested. I see people like to swap heater resistors to "fix their issues" but doing so without measuring the resulting voltage is kind of irresponsible IMO.


I'm looking now to see where I should be measuring B+, and what the correct level should be.
Can you share this? Quick Google didn't turn anything up, just stuff for the 2931.
Trying to help out @MoppelTheWhale a bit.
I didn't find it, but I am pretty sure you could just measure the voltage across the large filter capacitor in lieu of a proper test point. As for the correct voltage, I do not know.
 
@Hatsune Mike I figured as much, but without knowing the proper voltage there isn't much point.

I'll poke around on AO and see if grantspain can tell me.

Thanks though!
 
that could be the precursor to the flyback fully failing, that was seen on kortek digital chassis at the point of failure so it may be the same on this chassis
b+ 180vdc at the fuse near the hv regulation pcb
 
Back
Top