turbo grafx 16 repair

[chop5]

Howdy again,this is another one of my boring threads. This time for this non-working tg-16 ive had for years.
And like my other thread this one may take years to finish.

Every time I switch it on all i get is a solid blue or green color screen with no game in it.
With a game in it i get a different color solid screen like white or red,etc and sometimes a ping of sound depending on the game.
With a system card 2.0 in it i get a flashing screen with lines and beeps of sound.

When i plug in my working tg-16 without a game in it i get a solid orange screen.

Ive cleaned the cart slot like you wouldn't believe then checked every pin with the tester to see if there was contact. Plugged it in with a regular rf and ac adapter then with a cd unit with same results.
I'm not the original owner but im the first to open it. No other history is known.
Ive checked all of the components with a multi tester finding nothing wrong with them. No sign of melt or burnt smell.

So i went on a search for consoles with a similar problems.
If it was the crystal there would be sound and weird colors even black and white picture but game will load,
If any of the main processor chips were faulty there would still be video but all kinds of weird pixels and stuff,same if it was faulty voltage regulators.
What about cut traces or corroded traces? All traces were checked with a strong light and no sign of cutting or corrosion was found. The ram chips were checked for continuity to see if there was a bad trace. They were fine.
What about cracked or loose solder joints? All were inspected and are fine.
Might be a fuse somewhere inside that may have blown like the psx many fuse affliction. Theres only 1 fuse in the tg-16 and it checked out fine.
If the fuse is fine then how could a power surge or something not trip it saving the ram chips? I suspect energy feedback from a faulty controller and I found controller pins stuck in the controller port.
What are ram chips? Seems that consoles need the video and work ram chips to function properly
Idiot baka,why don't you trash it and just get another tg-16? Good question...

All i found with a similar problem was the neo geo aes. Its cause was bad video ram chips. Replacing them is the only cure.
So im gonna go this route and change the video ram chips on the tg-16. Only problem is finding the exact same chips or a possible substitute.

the 2 small chips side to side i think are the video ram,IC114 and IC110 printed with HSRM20256 LM12.
The chip under the shield thingy i think is the work ram,IC102 printed with HSRM2264 LM10
Both by seiko/epson with the two little triangle logo. surface mount 28 pin
I'm going by a specs faq i found for a super grafx that has similar chips and calls them video ram.

Does anyone know where i can find these chips? Or anyone have any info or pointers to add? Something
I may of missed to check?
(sacrifices a bull to summon dlite and other tg-16 guru's)
If i find and replace the video or work ram chips and nothing changes il change the voltage regulator chips next and go down the line replacing every chip till i fix it.








there are 5 more chips on a turbo grafx 16:
HUC6260-Video Priority Controller
HUC6280-the cpu
HUC6270-Video Display Controller
C358C-low-noise dual op amp
7805- 5v voltage regulator

some more board pics
http://img.photobucket.com/albums/v238/jetbue7/Pictures0lk01.jpg
http://img.photobucket.com/albums/v238/jetbue7/Pictures0y03.jpg
http://img.photobucket.com/albums/v238/jetbue7/Pictures0y01.jpg
http://img.photobucket.com/albums/v238/jetbue7/Pictures0y02.jpg
http://img.photobucket.com/albums/v238/jetbue7/Pictures00y7.jpg
http://img.photobucket.com/albums/v238/jetbue7/Pictures0y06.jpg

[Keranu]

Though I don't have any help for you whatsoever, I would like to comment that your pics are really nice!

[chop5]

Thanks keranu 

Some progress. Added some more info to the first post about the screen color and stuff.

The 2 chips are common and in many devices. Many substitutes out there.
With some help from channelmaniac over at neo geo forums hes located substitutes.
This is his quote:
They are both appear to be JEDEC compliant SRAMs.
Here's a replacement for the 256k one:
http://www.arcadecomponents.com/cata...46/4758714.htm
I don't have one for the 64k SRAM yet. I'm still trying to source one. A 6164 or 6264 SRAM IC in SOIC packaging should fit.

I will begin with changing the two HSRM20256 LM12 with the chips on that link and see if any change.
If no change and after my hulk like rampage il change the HSRM2264 LM10
NEC Gods help me.




heres some tech info on both chips:
For the HSRM20256 LM12. Still haven't found the official pdf or document for it:
http://img.photobucket.com/albums/v238/jetbue7/256.jpg

For the HSRM2264 LM10:
http://img.photobucket.com/albums/v238/jetbue7/64.jpg
Pdf can be found here:
http://www.datasheets.org.uk/search.php?t=0&q=SRM2264+LM10&manystr=&sub.x=34&sub.y=14

[Platinumfungi]

Nice work! I'm curious to see how this turns out...

[nodtveidt]

I can tell you with 100% certainty that the 7805 you see is indeed a voltage regulator. It is designed to keep a steady 5V flow from a higher voltage source. The C358C is a low-noise dual op amp, it's NEC's version of the LM358.

[chop5]

Ah thanks The Old Rover . Il make the correction change.


I haven't changed the chips yet cause I'm being told by others that the chips may be fine and could very well be a hucard slot issue. That before i start changing stuff i should attempt a slot change or transfer or at the very least solder wires from the points to a cheap game to see if theres contact.
I will attempt this while learning SMT soldering for the chips if this slot transfer doesnt work.
Lots of vids on you tube for SMT soldering and chip removal. I don't have a fancy schmancey soldering station so all il be using is a regular iron. The instructional vids on the tube seem easy enough and il practice on many junk boards i have here.
Or get someone with a SMT soldering station to do it for me if i get frustrated and cant do it 
Anyone here with a SMT soldering station?

[l_oliveira]

2x 32kbyte (256kbit) are the VRAM
the 8kbyte (64kbits) are the work ram and is on the small board with the hucard slot.

By what you said on the very first post it seems like the CPU is not able to fetch instructions from your hucard. (since it's a US unit I would say turbochip instead lol) but indeed I would check and re-check the hucard slot as a very first step. Maybe someone tried to clean some cards with steel wool sponge. There would be lints inside the connector and it would be enough to cause malfunction.  Trust me I repaired SNES and Super Famicoms with such problems more than once.

[chop5]

Ready for a good chuckle? Gentlemen behold.....my hucard test slot!







Used a slot from a dummy board and cut it to size took many hours of soldering. 
i couldnt figure out a easier way to do it.

 The only thing that bothered me is i had no way to test it on another tg-16 cause the other unit i have is hard modded to japanese and doesn't even have the spokes to solder on anymore like this one. I checked all the wires with a meter,but one sure way to make sure its working was to put in the system card. When i put the system card in the actual slot i get a kaleidescope of colors and sound.

All the games i tried had the same result if i was using the actual card slot. Different color screen every time i turn it on and sometimes a ping of sound. Then i put in the system card and got the many colors and sound. So it seems it works but that means its not a slot issue and il go on to the next phase and change the work and video ram chips.

Anyone have any ideas how else i can test the slot than this abomination i have made?

[Michael Helgeson]

If you get it working,reverse some wires and have  a dual slot system,no switch needed

[SignOfZeta]

I hope you really enjoy this hobby, because you've already put way more work into this system than what it would be worth if fully functional and mint condition ($20).

[nodtveidt]

Some people enjoy tinkering, some people enjoy doing it for the educational value...understanding first-hand how the console works...or in this case, worked.

[ceti alpha]

Nice work Chop5.  Yeah, sounds like you're having fun tinkering around with this stuff. I hope you get it working. 

[sinfreealex]

I've got to say this is the greatest thread I've read in a long time. I restored arcade machines in the late 90s and was shoulder deep into this kind of stuff all the time. I've been missing it all these years and this post makes me feel all warm and fuzzy inside. Please keep updating when you can. I'll be here watching with baited breath. 

[chop5]

I just sent the unit last saturday to arcadecompnents.com to have the chips professionally removed and replaced.
Now we wait for word or its return.

[channelmaniac]

Hey everyone,

I found the site searching Yahoo! for my user ID.

I changed day jobs and finally got caught up to the point I got the system on the bench.

I pinned out the CPU a bit (address, data, power, reset, clock) and the HuCard socket from the pinouts on the Work SRAM IC.

Next I checked the reset line and it wasn't being held low long enough. Replaced the capacitor on the RC network feeding the reset line. Reset was working better but was not the cause of the problem.

There I found a bad trace on address line A2. Patched that but was getting strange signals on the SRAM with the logic probe. Replaced the IC. Signals were better looking but system was still not booting. Cleaned the card slot (folding rough paper over the end of the card, wetting it with pure alcohol, and feeding in/out repeatedly) then resoldered it. There were quite a few solder connections on the connector that were cold - they bubbled when heated - which had to be desoldered and resoldered.

Game system is now working.

Raymond
arcadecomponents.com