Jump to content
IGNORED

130XE brought back from the dead - Advice re glitching display


Beeblebrox

Recommended Posts

2 hours ago, Beeblebrox said:

Yup I was gonna focus on the transistors - just need to research how to test them,

It's difficult to test transistors in-circuit, you usually have to remove them, then testing is easy.

Put meter across Emitter and Collector set to either resistance (or diode test if your meter has this function), your should get open circuit.

reverse the connections and again open circuit.

If it's an NPN transistor, place +ve meter probe on the Base and in turn place the -ve probe on collector, then emitter

you should get a resistance reading both ways.

Now put the -ve terminal on the base and again in turn the +ve probe on collector and emitter

you should get open circuit.

 

For PNP the testing is the same across collector/emitter.

but reversed for base to collector/emitter

 

The video circuit appears to have two NPN transistors - 2N3904

here's the pinout

image.png.88d366022796f7c5589adb9f50779331.png

Hope this makes sense.

Link to comment
Share on other sites

Looks like a PAL machine, since W2 isn't jumped.  Isn't part of the video circuit in the RF modulator, I don't have a modulator (remake board), and use UAV and VBXE, so I can't help with that. 

 

I'm assuming this isn't the case, but if the monitor is an NTSC monitor, it will only display black and white from a PAL machine.

Link to comment
Share on other sites

ok another update.

 

Just to confirm I just put a precision round socket in place of the old one for the 4050 chip. I then took the 4050 and 74LS74 chips from this 130XE and popped them in my 600XL, they both worked fine. I took the 600XL's 4050 and 74LS74 chip and popped them in the 130XE and still have the black and white display. Whilst I was in there I took the 130XE GTIA chip and popped it in the 600XL and got colour.  So at least I know these three chips themselves don't cause the issue. 

 

@TGB1718 thanks for the info - it's looking increasingly likely that I may have to go down that route. I'll digest you post and figure it out. 

 

@thorfdbg and @wildstar87  Yes I am UK based and this is a PAL 130XE running on an LG Flatron M227WD using either a DIN to S-video or DIN to Composite. All my other PAL A8s run colour with both of these cables (All my XLs have had their Chroma connected up). 

 

@thorfdbg "the color clock is a separate crystal that feeds GTIA, so you may want to measure whether the color clock arrives at GTIA."

Can i ask where does the colour clock signal come into the GTIA just out of interest? Is there a specific line in through some components and/or into pin(s) on the GTIA? If so which pin(s)?

 

Generally to sum up so ya'll don't have to read the first few posts:

 

So far:

  • Dead PAL 130XE rev 4 board had no picture (black screen) although monitors and TV recognised a signal when it was switched on. 
  • My trusty Sys check 2.2 couldn't help this time as no image.
  • Eventually I worked out the main reason was bad Micron MT RAM causing it not to boot. (Well of course it was!!):roll: I discovered this after undertaking the arduous task of unsoldering, removing and testing all 16 MT chips.
  • I socketed and replaced all RAM chips with Samsung brand chips, testing each new chip before installing it on the board with my DRAM tester.... and it finalyl showed life and booted to Ready prompt.:)
  • Happiness was shortlived and straight away it was apparent there was a graphical glitch with a grey distorted vertically lined display and an increasingly jumping image.
  • Eventually I found pinhole 2 on the PCB under the ANTIC on the underside of the board had a bad via connection so fixed it by jumping a wire to the point it should connect to. 
  • Instantly the image was restored and no more line or jumping glitches BUT still had a black and white display only on both composite and S-video cables. Both cables work with all my other XE/XL machines and give colour with no probs.
  • I've swapped out and tested ALL chips and it isn't the chips that are bad. (ALL RAM chips were checked individually prior to installing them. GTIA, ANTIC, CPU, MMU, FREDDIE, 4050, 74LS74, PIA, POKEY, OS, 74LS08 (replaced with 74F08 and several of the latter tested). All chips have gone into working machines and visa versa and work.
  • All sockets for FREDDIE, MMU, ANTIC, GTIA, CPU, 4050, 74LS74, POKEY and PIA check out with some replaced with precision sockets as  precaution and then rechecked. All contact points from sockets traced beyond the socket legs themselves to other contact points along traces to other point using my multimeter.
  • I've also eyeballed and tested lots of other solder points on the board generally.
  • I replaced the PAL clock (CO16801) and the crystal with other indentical from a donor board - made no difference - (although they are from a donor 130XE rev 3 board so I can't 100% say they are reliable components obviously)
  • Fixed an unrelated issue with the keyboard ribbon connector so keyboard works.
  • To confirm Pokey and sound is working fine. I played the Drunken Chessboard demo with no issues aside in black and white!
  • Cart port works and my SIDE3 cart boots no issues.
  • The basic chip is soldered directly to the board and I've not touched it - primarily because I believe it is incredibly unlikely it has any involvement in this issue.

 

Still to do:

 

  • I may replace the CPU socket with a precision socket. Holding fire for now as it looks ok and I don't wanna tempt fate.
  • On advice from others I am currently looking into the trim pot voltage range, (and I have a replacement on order just in case),  and it looks like I may have to start looking at the passive components, in particular transistors. Looking at the passive components looks a big job and I need to gather as much info as possible and re-read some of the replies I have in this thread already (thanks all), before I undertake it.

 

I also have a replacement 74LS74 on order - although as mentioned above since ordering I've checked and it isn't that chip. The socket for the 74LS74 checks out ok too.

 

I think that covers it.:-D

 

So just to say it is definitely nothing external (ie cables, monitors). I believe it is something in the PCB's video display circuit or related. 

Edited by Beeblebrox
Link to comment
Share on other sites

1 hour ago, TGB1718 said:

It's difficult to test transistors in-circuit, you usually have to remove them, then testing is easy.

Put meter across Emitter and Collector set to either resistance (or diode test if your meter has this function), your should get open circuit.

reverse the connections and again open circuit.

If it's an NPN transistor, place +ve meter probe on the Base and in turn place the -ve probe on collector, then emitter

you should get a resistance reading both ways.

Now put the -ve terminal on the base and again in turn the +ve probe on collector and emitter

you should get open circuit.

 

For PNP the testing is the same across collector/emitter.

but reversed for base to collector/emitter

 

The video circuit appears to have two NPN transistors - 2N3904

here's the pinout

image.png.88d366022796f7c5589adb9f50779331.png

Hope this makes sense.

@TGB1718  Just to say there are 4 x 2N3904 448zf NPN transistors on the 130XE rev 4 board circled in Green on this annotated pic below.

 

Would you suggest checking all 4 x out by removing them as per your instructions above?

 

What output am I looking for (sorry to be a novice) ?  I googled 2N3904 448zf NPN transistors and various different ratings popped up.

 

 

130XE circuit board annotaed image with 2N3904 transitors circled.jpg

Edited by Beeblebrox
Link to comment
Share on other sites

4 minutes ago, TGB1718 said:

Your only interested in Q3 which provides luminance (probably ok)

and Q2 which provides Chroma.

 

Don't know if you've got these, Page 3 has the video circuitry on it.

Schematic_130xe_Jerzy_Sobola_Landscape.pdf 243.02 kB · 1 download

@TGB1718Great, thanks for that. Time to go to school as I am a real beginner at reading schematics - :-D

 

I also came across this as a general parts ref: http://www.atarimania.com/documents/130XE-Reference-Manual.pdf for anyone who is generally interested

Edited by Beeblebrox
Link to comment
Share on other sites

1 hour ago, Beeblebrox said:

@thorfdbg "the color clock is a separate crystal that feeds GTIA, so you may want to measure whether the color clock arrives at GTIA."

Can i ask where does the colour clock signal come into the GTIA just out of interest? Is there a specific line in through some components and/or into pin(s) on the GTIA? If so which pin(s)?

It's pin 16, PAL. The signal is generated by U12 and the crystal Y2. You should also check the pot R38 connected to pin 17 of GTIA "CADJ". Pots are mechanical parts and they age. Probably try moving it foreward/backward, and check whether some colors appear.

  • Thanks 1
Link to comment
Share on other sites

22 minutes ago, thorfdbg said:

It's pin 16, PAL. The signal is generated by U12 and the crystal Y2. You should also check the pot R38 connected to pin 17 of GTIA "CADJ". Pots are mechanical parts and they age. Probably try moving it foreward/backward, and check whether some colors appear.

@thorfdbg  thanks. Yeah I know the pot was reponding as the voltage does change when reading with a multimeter. You can tell on screen it's being turned although still a black and white image whatever direction I turn it.  I am currently cross referrencing the schematic TGB1718 provided against the 130XE ref manual as the latter has an albeit poor scan of the part numbers as they sit on the board (from page 10 onwards). 

 

Thanks for the heads up re pin 17 of GTIA. I am also tracing pin 21 of GTIA to see which components it comes across as they sit on the board. Schematics are very new to me and whilst they are starting to make sense they can't tell me where to look on the board. (Most of the passive component's numbers are obscured (naturally) by the components themselves).

 

Edit: this PDF is helpful also from page 8 (attached) as found on this post on AA

 

Although U12 appears to be pointing to a ram chip? 

 

130XE - Sams ComputerFacts.pdf

Edited by Beeblebrox
Link to comment
Share on other sites

ok, so I extracted and tested the Q2 2N3904 448vf marked transistor (positive probe on base, negative on collector). I don't really know what voltage level I was expecting to see but I get 0.900v

image.thumb.png.21fb955835343d9e9d8fd7cd08fdd5f3.png

I googled and got this:

For an good NPN transistor, the meter should show a voltage drop between 0.45V and 0.9V. 

 

So I guess it means the transitor is ok. (Again I am a novice at this).:dunce:

 

The Q3 transistor reads 0.855v:

 

image.thumb.png.c7127dd247d4fbe1dca648936a9e41cd.png

Link to comment
Share on other sites

Ok the latest before I quite for the day.

 

After testing the 2 x 2N3904 transistors and them appearing ok, I popped them back in.

 

I then decided to remove the trim/colour pot from my donor board, take some readings on my multimeter, clean it up with contact cleaner, and install it in the 130XE. 

 

So I remove the one in the 130XE (which is white), cleaned it up, took some readings, then put it aside to install the doner one (which is green).

image.thumb.png.2914a414e0d8506e5910087413836bfe.png

 

With the removed (white) original 130XE trim/colour pot I was getting readings starting 450ohm and going up to 2.5v as I turned it:

image.thumb.png.08ddb8f531bd95d57771eac84117adb7.png      image.thumb.png.a7f3422ba8cfc97a66594e86afdb0d9b.png

 

I double checked the newly installed donor trim pot connected to the right points from the vias and traces. All good.

 

Powered up the 130XE and same grey screen.

 

At this point I dial the trim pot down anticlockwise a mm at a time. Nothing really noticable until a quick glimmer of pink and then gone.

 

It's only a small fraction of the scope of the dial but you get this if you go slow:

 

image.thumb.png.d0e29a947ecbe4da7389d86c055da985.png

 

...move it half a mm and you get this:

 

image.thumb.png.4d2403ff78162d7bfebdd23f887c06d5.png

 

and then move it half mm more and back to the grey screen.

 

So whilst I can't rule out the original trim pot had this and I never noticed it because it is incredibly easy to miss it without realising when turning it, it is insteresting nonetheless.

 

Anyway, I have a new trim pot on order so will replace the donor one. I don't think the trim pot is the main cause of the lack of colour.

 

I still think something else is causing the issue in the video circuit and will have to continue studying all the schematics and absorbing all the advice given in the thread.

 

Thanks again for all help so far.

 

 

Edited by Beeblebrox
Link to comment
Share on other sites

5 hours ago, wildstar87 said:

xefsm.pdfThis is a version of the FSM that someone updated to modern standards, pretty nice.  There is a troubleshooting guide at the end, it does list possible components, including the RF modulator.

@wildstar87 thanks so much for bringing this pdf to my attention. I had no idea there were quite so many XE board revisions. Like you say this section is particularly interesting :

 

SECTION 4 
SYMPTOM CHECKLIST

No Color or Bad Color Bars VR1 Adjust, U20, U2, U19, 
Q3, Y1, RF Modulator
 
I think I've looked at U20, q3 and y1. 
 
I am not sure I'll need to look at the rf modulator as I am not hooking the display up via this. 
I'll look at u2 and u19 I think. 
 
Thanks
Link to comment
Share on other sites

Ok, few hours spent this morning as follows - no breakthrough as yet :?:

I spent this morning following the trail for possible issues with U2 and U19 chips (see post #43 above):

 

Removed the U19 (555) 8pin chip, installed a socket:

 

image.png.57c71f966ced68523eec82cca750e381.png   image.png.f7f5ab3113df53b9f66b4a8ee61a49b5.png   image.png.00cc36fb7dcda0bb4d8fe9ddec503934.png

 

Removed a known working U19 chip from another XE, popped it in the socket.

image.png.6192df5a2671aa43a305d27a28c6b81c.png

 

Powered up the machine - No change - black and white screen. Turning trim port has same behaviour as before (mostly just the grey screen and then the pink one detailed last night a few posts ups).

 

Took the U19 chip I'd removed from my 130XE and popped it in my other XE to test - no issues . So we can rule out the U19 (555) chip itself.

 

Then did exactly the same for the 74LS138 U2 chip - no change and the chip works in my other XE:

 

image.thumb.png.30091a39fc496716703ce6ece900edca.png   image.png.8e43121f6ec1c20db8f5fea9b2c77472.png   image.png.eebf12d04138701739485559fbadf1a8.png             

 

I also took another 74LS138 chip I had and tested it in the 130XE just in case. No change. So we can rule out the 74LS138 chip itself.

 

FYI For both chip's new sockets I checked that all connection to legs and points along traces all worked and solder work was ok. I'd also cleaned up around the sockets and board both sides, etc before testing.

 

Whilst I was in there I decided I'd socket the U5 28pin OS chip, adding a precision socket. Checked it all out after installing and all ok in that respect. (Will come in handy for the U1MB upgrade if I every get this colour issue sorted):

 

Old socket:

image.png.41d2d33c1fc800295943563b32f82ca9.png

 

New socket:

image.png.f3e87691bfa6dba9d5d45c500056ef3c.png   image.png.6c7245cf96f664e60c4d152a410f6665.png

 

Again I cleaned up before testing.

 

What I also did was check the reading on the trim pot (when XE off) whilst the trim pot was set to the position where I was getting the pink screen as detailed last night a few posts up.

0.940V is the reading.

 

The reading is taken rom the right hand side contacts, which I presume is how you take a reading with these particular potentiometers?  Certainly it was the only side that registered any meaningful output.

 

image.thumb.png.35ea02047c1693153f1e9fb673101d6c.png

 

So this is the board as it stands:

 

image.thumb.png.74bb465438cba45f990acf0c29a0e0f7.png

 

BTW yes if you look at the GTIA chip it does look upside down.:grin: This is just the way the markings on the chip were printed in the factory. (I've seen this from time to time). The chip is definitely installed the correct way and also I've previously had other known working GTIA chips from other A8s in place and still the black and white issue persists. (Just thought I'd mention it). :)

 

Back to reading the advice in this thread, absorbing schematics and field manuals. Gonna check a lot more of the passive conponentry and the PCB traces, vias, etc etc. Long weekend ahead methinks. :roll::P

 

This has been very useful:

 

image.thumb.png.c1341da2783e875801f8b4def6e9cafb.png

 

Thanks again @wildstar87 @kheller2 and @TGB1718 for PDF docs - they have been invaluablem and thanks to all who have pitched in with advice.

 

Edited by Beeblebrox
Link to comment
Share on other sites

I am just going through the components along the lines that connect to pin 21 of the GTIA.

 

I noticed when comparing Jerzy's schematic (2003) that there appears to be an error with R201 in two places. This of course could just be me misunderstanding the schematic but there are 2 x R201s with different resistor values:

 

So here on page 3 of Schematic_130xe_Jerzy_Sobola_Landscape.pdf

 

Jerzy's 130XE 23003 schematic with 2 x R201.jpg

 

and on the SAMs computer facts pdf schematic on page 3 also:

SAMs computer facts 130XE schematic with R201 highlighed.jpg

So on both the pin 21 goes to an R201 6k8 (6800ohm) resistor which I found (and replaced on the board (made no difference btw) - but Jerzy's schematic shows another R201 valuesd at 5k6 (5600ohm) which goes to the L32. On the SAMs schematic I can't see any reference to the L32 or another R201.

 

Unless of course the schematic has R201 twice because there are two connections to it. Then why does R201 have two different resistor ratings?

 

So far I've replaced R201 and R205 on the board with the same value resistor in each case. - made no difference and the original resistors removed checked out ok.

130XE - Sams ComputerFacts.pdf Schematic_130xe_Jerzy_Sobola_Landscape.pdf

Edited by Beeblebrox
Link to comment
Share on other sites

Check the circuit that takes O1 from Sally (6502CPU) and creates ColorAdj if you haven't done that yet. I cut a piece of the SAMS schematics and it is attached. It basically takes a pulsing signal in TTL level and increases the voltage to 12V (or 9V... don't quite remember).. anyway, check if you can see this higher voltage which is the one that finally is trimmed by the potentiometer. I once kept trying to fix the pote and in reality the pote was good, but the voltage to the pote wasn't there: the O1 signal going to that circuit was not there because there was a connection issue in the Sally socket.

Just my $0.02

 

image.thumb.png.a1c4baccb860103e8b1c6fcecaa6f4c6.png

Edited by manterola
  • Thanks 1
Link to comment
Share on other sites

Keep in mind that both Atari and Jerzy's schematics probably have errors, and that both probably have used the same component number more than once.  In fact, Atari has done this on a few occasions even on the silkscreens.  SAM's also has errors on the video jack pinout, as I recall.

 

Take a look at the stock video circuit as drawn in the Super Video for the 130XE docs.  The RF MOD provides composite out, which is why you can't remove the modulator.  I know you don't use the RF, but have you tried connecting that to see if you get color?

 

.349124328_ScreenShot2021-09-24at10_24_30AM.thumb.png.bcdd2ed0969d45152e142440665e2984.png

  • Thanks 1
Link to comment
Share on other sites

3 hours ago, kheller2 said:

Keep in mind that both Atari and Jerzy's schematics probably have errors, and that both probably have used the same component number more than once.  In fact, Atari has done this on a few occasions even on the silkscreens.  SAM's also has errors on the video jack pinout, as I recall.

 

Take a look at the stock video circuit as drawn in the Super Video for the 130XE docs.  The RF MOD provides composite out, which is why you can't remove the modulator.  I know you don't use the RF, but have you tried connecting that to see if you get color?

 

.349124328_ScreenShot2021-09-24at10_24_30AM.thumb.png.bcdd2ed0969d45152e142440665e2984.png

@kheller2thanks

I'd not thought of seeing if rf worked. Will try. Yup I figured there might be errors in Jurzy's schematic. Not Atari's though... Interesting. 

 

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...