Need assist with K4915 chassis

SuperSprint

Well-known member
Joined
Jun 29, 2010
Messages
2,753
Reaction score
664
Location
Georgia
My 19K4915 chassis (Super Sprint - Atari) died after many years of fun. I've got a basic electronics background, but no practical experience with monitor repair. I'm hoping some of you vets will be willing to lend some sage guidance about how best to proceed.

Problem: Sudden failure. No picture. Game still playing "in the blind".

Testing: No neck glow. HOT (Q352) has low impedance to ground. 3.3ohm resistor in primary voltage circuit (R601) tests open. Some signs of cold solder, but continuity tests to far points suggest they are good. back right of PCB shows some signs of darkening/cooking, but no obviously burnt components.

Notes: The flyback looks good (no cracks or burns), but when I took it to a tech several years ago he diagnosed it as a bad flyback and said I needed a new monitor. He may be correct, but I felt "dismissed". Having since read that flyback failures are rare in this monitor, I'm unsure of this diagnosis.

What I have: Cap Kit, schematic, PCB prints, solder iron, DVM, Fromm flow chart.

What I *think* needs to be done: Replace R601, Q352, all electrolytics, pray.

Anything else I should consider before digging in?
 
Any risk of "circular failures" or components that I should replace no matter what? For example; I replace the HOT and R601 but the skip the VR IC and cook HOT&R601 again.
 
Typically the only parts that cause cascading failures are the HOT and flyback. Do you have access to a ring tester? You could verify the flyback is good if you do.
 
I would definitely check the regulator for shorts. If the flyback is bad you'll probably blow another HOT, but I don't think I've actually ever seen a bad 4900 flyback that didn't function (but I've seen several that make a loud whining noise). While you're working on it, keep a close eye out for bad solder joints and resolder anything that looks suspect.

Word of warning, in case you didn't notice, arcade monitors are "hot chassis". They require an isolation transformer to function. Never plug it directly into the wall. You *will* damage things if you do. The isolation transformer is a 1:1 winding on the main power transformer on many games, or a separate transformer on others.

-Ian
 
If it makes any difference the main chassis fuse F501 is good and I do not recall it having ever blown. Interestingly, AC voltage at any point except the line cord solder points (i.e. across C501) is about 69V. At the solder points it's nominal line voltage (~110V)

Typically the only parts that cause cascading failures are the HOT and flyback. Do you have access to a ring tester? You could verify the flyback is good if you do.

Since I don't know what that is, I'll say no.

I would definitely check the regulator for shorts.

Given the price and ease of replacement, would it be bets just to replace it for good measure?

Word of warning, in case you didn't notice, arcade monitors are "hot chassis". They require an isolation transformer to function. Never plug it directly into the wall. You *will* damage things if you do. The isolation transformer is a 1:1 winding on the main power transformer on many games, or a separate transformer on others.

I knew that, but I'm not aware of all the implications of such an arrangement. Is there a good post on the topic? I'm presently using the game itself for monitor power.
 
Last edited:
But it's a tool, right? :)

Looks like Chad now has a min order of two flybacks. Is that flybacks only?
 
That ring tester sure is a cool bit of test equipment, but considering you can buy a brand new flyback from Chad at arcadecup.com for $30, I think that the ring tester might be overkill for fixing one monitor.

:D

-Ian

Except that you may have a good flyback already, and then you just wasted $30 on a flyback. With a ring tester, you can test others later before spending money. I like mine...
 
I'm stuck with this one, and any help would be greatly appreciated.

Before;
- No picture, no neck glow.
- R601 open
- HOT collector-> ground shorted.

Here's what's been done;
- Capped it and verified polarity (except auto-protect board)
- Reflowed any suspect joints.
- Replaced R601 (3.3ohm) between rectifier circuit and voltage reg circuit.
- Replaced IC301 (STR380) voltage reg with STR381 (sub for 380 per Zanen)
- Replaced HOT (2SD870).
- Verified R601 = 3ohm
- Verified R503 = 180ohm

Here's what happened;
- No picture, no neck glow.
- R503 got really hot and smoked a little.
- Fuses remained intact

Test Data
- TP51 (Vout from rectifier circuit, before R601) = 145VDC
- TP52 (Vin to AVR circuit, after R601) = 145VDC
- R503 in/IC501 pin1 = 145VDC
- R503 out/IC501 pin2/TP91 = 0VDC
- Collector of new HOT is shorted.
- Discharging anode after test yielded an audible snap.


Fromm Chart says: IC301 (voltage reg) failure.
 
Last edited:
PART 2 - Concerned that STR381 might not be a proper substitution for STR380, I shut down and reinstalled the old, untested STR380

What happened;
- No picture, no neck glow.
- Power supply main fuse blows.
- Monitor chassis fuse (same size) remained intact
- Discharging anode after test did not yield an audible snap.

Test Data;
None. But before the "repair" it just played blind with no hot components or fuse blowing.

--------------------------------------------

PART 2 - Since STR380 is clearly not working, I replaced the main fuse and reinstalled STR381.

Here's what happened;
- No picture, no neck glow.
- R503 got really hot
- Fuses remained intact

Test Data
- R503 & R601 proper spec when tested in circuit.
- TP51 (Vout from rectifier circuit, before R601) = 145VDC
- TP52 (Vin to AVR circuit, after R601) = 145VDC
- R503 in/IC301 pin1 = 0VDC
- R503 out/IC501 pin2/TP91 = 0VDC
- Vdrop across R503 = 145VDC
- Also tested with STR380 removed, same effect
- Discharging anode after test did not yield an audible snap.

--------------------------------------------

So that's where I'm at. Something is happening in the second half of the AVR (post rectifier) that wasn't solved by a new reg IC. Looking at the schematic the common component between the HOT and IC301 is the flyback...
 
Last edited:
Fromm Chart says: IC301 (voltage reg) failure.

Actually, the flowchart says IC501 (STR380), but you might have made a typo.

First doublecheck the wiring to your VR to make sure you didn't hook it up backwards. Also, did you put the mica insulator between the VR and the side wall?

If you have zero volts DC at TP-91, then either IC501 is bad or miswired, or R503 is bad - or both.

And go through and doublecheck the solder again. Any part that is taller than the fuse should be wiggled while watching the solder pads below. Some pads will look okay but still have problems....
 
Actually, the flowchart says IC501 (STR380), but you might have made a typo.

Yes, thanks. IC501.

First doublecheck the wiring to your VR to make sure you didn't hook it up backwards. Also, did you put the mica insulator between the VR and the side wall?

Mica = Check. Backwards isn't possible as it's a TO-3 4pin package.

If you have zero volts DC at TP-91, then either IC501 is bad or miswired, or R503 is bad - or both.

T503 tests at 180ohms (spec) in circuit. I can check the pads at IC501, but it's a factory socket that hasn't been modified since the unit was known working. I'll have to get a new STR380 as I don't trust it anymore.

Looking at the schematic, when I tested it with IC380 removed I should have seen voltage at the R503 input (and output [TP91] I would suspect). Yet none at either point.

So even with it killing HOT's and IC501's, you don't suspect the fly?
And go through and doublecheck the solder again. Any part that is taller than the fuse should be wiggled while watching the solder pads below. Some pads will look okay but still have problems....

I'll do that. I did a continuity test from pad to pad on many that looked suspect, and reflowed 'em to be sure. Now is when I REALLY want a pattern generator as I do not like using my SS PCB as a test rig...
 
Last edited:
I didn't see you saying that it was still blowing the HOT, too. Probably is a good idea to do the flyback at this point just to be sure. If you happen to have a good 4900 lying around, try swapping it in to see if that's the problem...
 
That was my thought. I bought two HOT's just in case, but not two STR380's. Man those little buggers are expensive for what they are... Sadly, no donor chassis so I'll get one (or 2) from Chad.

Also, you guys mentioned testing/adjusting the B+ voltage quite a bit in other threads. That particular label (B+) isn't not marked on my schematic or referenced in my manual. Is that the voltage leaving the power supply secondary circuit, feeding into the flyback (TP91) or the input at the flyback at pin 9?
 
Thanks, but was B+ at TP91 or pin 9? :)

Both. If you look at the schematic, you'll see that TP91 and pin 9 of the flyback are on the same trace.

But it's usually recommended to measure it at R503 on the side closest to the fuse, as it's usually easier to get to when the monitor is installed in a cab...
 
Back
Top Bottom