Amiga 1200 color problem repair
Вставка
- Опубліковано 20 лют 2013
- This Amiga 1200 motherboard had some sort of color problem. During some testing, I noticed that the only color that showed up correctly on screen was the green one. Both the blue and red color was behaving improperly as I was moving the color slidebars for each color respectively in ProTracker.
I have here documented the repair process of this motherboard, including the steps of initial testing and problem observation, circuit analasis and diagnosis in order to track down the problem, futher testing for fault verification, and finally replacement of the components.
On the very same motherboard, there was also a problem with the audio which I figured out was not related to this board in particular, but most probably a problem on all A1200 rev 1A motherboards. Later revisions are not affected.
However, this video is dedicated only to the color problem. I made another video regarding the audio problem, so feel free to have a look at that one as well. :)
• Amiga 1200 rev 1A audi...
Questions and comments regarding the repair of this color problem are always welcome. - Наука та технологія
Yep, exactly. This is a 8-bit video DAC, meaning it is working with 8 bits for each color respectively, with a total of 24 incoming bits for the video. In the datasheet for the video DAC, it is also explained that R0, G0 and B0 are the least significant bits, climbing up to the 7's which are the most significant ones.
I'm impressed with how quick you were with your comment after I posted the video, btw. :)
awesome video mate, love how you diagnosed the issue from the components to the pulse to identify the video dac was the issue , thanks it was great to learn more from you.
Ob btw, and the way that it turns to black in the middle, indicates that the next higher bit is not working: example with 4 bits:0000 0001 0010 0011 0100 0101 0110 0111 then it should be 1000 but it's 0000 and starting from beginning. So the highest bit was defective. This is an alternative mathematical way to figure out which pin is not working (but not at which chip)
This came to my mind too right as soon as he playing with the sliders. And then when he was playing with the probes you can see that number of times it goes high is 2, 4, or 8. And the most significant bit would have been only 1 time had it been working... off or on.
I found this video one of the greatest I have ever seen. So far I didn't know about your work and it seems that I will stay a little on your channel. Greetings from majsta.
Imajstorovic Thanks! I'm glad you find the videos interesting.
A fantastic in-depth tutorial. A really interesting project well done on getting the video working correctly. I have a logic probe but after seeing this I feel I must now invest in a logic pulser also.
Thanks for your encouraging comment. Appreciate it. :) I can not perform surgery on flesh and bone, so I hope your life is not at stake. However, I often manage to cure patients who mainly uses electricity as their main nutrition. :)
I would be happy to perform an operation on your A1200, if she can only afford to wait a little bit. I have lots of other hardware I need to have a look at before I can accept new patients, and I only have so many hours a day for electronic repairs.
I just want to say that your channel is amazingly fascinating
Thanks a lot.
Yes, you are correct. The video DAC on the A1200 is a 24 bit device (8 bits per color). The bits are named 0-7. In this case, it was bit B7 that was not working, which is the most significant bit for the blue color. In the middle of the slidebar, it should have changed to 1000 0000, but instead it changed to 0000 0000 since there was a problem with B7. That's why it went back to black in the middle of the slidebar.
Using the pulse injector was a good way to demonstrate this however ;) The binary logic seemed clear when you thought B3 or B7 were at fault and using the evidence it could be derived that the HSB was at fault as it is only at level 1 twice in the full range (bit late in the day for my comment, no pun intended ;))
Another great performance of your certified fix-it magic :)
Actually my 90s Amiga had the same "everything went green" symptom. I can't tell if it was the same cause of course. But it happened over night. I didn't swap any resistors :)
Interesting video, thanks for sharing the knowledge. My A1200 also loses some red color and everything becomes slightly cyan when I screw the top cover too tight. It seems that loosening the screws fixes the problem for me. I hope it won't get worse over time.
Another informative entertaining vid my friend you sure have a talent,can we please have your animated helper back aswell in the next one, he was a blast !
gracias!!! he aprendido un montón!!! thank's i learn very much!!! amazing. Hi from Mallorca Spain
Excellent video. Many thanks. Loved watching this all the way through.
Cheers
Jim
The schematics for different Amiga models are actually very easy to find. If you just google for "Amiga schematics" or something, I'm sure you will find several sites that have them. I just picked a site a while back that seemed to have schematics for many different Amiga models, and I downloaded them all and stored them for future reference.
I really enjoyed this vid. I'm a bit jealous that my A1200 is working now. I have two dead ZX Spectrum +2's to keep me occupied in the meantime though :). Like and Subbed.
exactly! This is an alternativ way which works with colors quite well and simple arithmetic malfunctions.
Just about the time! good job! BTW do you have a service manual or a book or something? where do you got the schematic that you are using in your videos?!
Thanks for your comment. I'm sorry to hear about those strict regulations. I didn't know this about Argentina. It must be tough to not be able to freely buy any kind of electronic equipment you want or need from abroad. I wonder what the reason is behind these strict import restrictions? What do the they claim to be the reason?
I hope you can buy an A1200 and a C128 sometime in the future. Let's hope the government of Argentina lowers their resitrictions a bit in the future.
Very pertinent and presented in a very entertaining way! I have an Amiga 600 with a similar problem. The colors suddenly became tinted with bright yellow. The funny thing is that the problem arrived overnight so it's not a cheap repair from the previous owner. On boot up, the floppy reads a few tracks and then stops while the screen displays the classic boot screen. So it's not currently possible to load some utiliites to help troubleshooting the problem. Looks like the Amiga can not recognize any floppy as bootable. Can I send you the motherboard and how ?
thank you that was a great video, learned a lot.
Beautiful indeed - I absolutely enjoyed watching this one. Interesting to see the behaviour of the 2 swapped resistors, but I especially enjoyed the 2nd part. Thanks for showing the Logic Probe and the Logic Pulser - really nice to see properly how things really work. Would love to see once how you soldered the PLCC (VideoDAC), maybe will see an example how you solder in one of your further videos (as I'm going through them one by one). That jumperwire was interesting... which motherboard revision was this? I have a Rev 1D.4 infront of me (also with the VP101-3BA), but without jumperwire...
Ops - just rewinded to the beginning of the Video and saw that it was Rev 1A you had infront of you
BTW what software you are using to edit your videos! Looks very professional!
It's difficult to say a price, since I don't have any set prices. You will of course have to cover all shipping fees, all my expenses for spare parts etc, and a little extra for my time. I put the time needed into every repair and never do a hasty, sloppy work. Because of that, I never charge per hour because it would be too expensive for others to let me fix their stuff. Enough for me to feel there is a point in doing it, and low enough for you to also feel there is a point. That's my price. :)
Hi, first of all, you rock :)
I have a question, I don't know if I missed it but I was rewinding the video looking for explanation why the video DAC has two pins bridged via yellow signal wire, and I don't know if you're elaborating on that or is it a normal situation.
EDIT: I'm too impatient, sorry :) I wrote that comment before watching the whole thing
Yep, it's true that I didn't really go into the technical details about why the wire is needed on the chip. But it's all very simple, so I can answer that question here instead:
One end of the wire is hooked up to pin 16 of the chip. The other end is connected to pins 20 and 21. Pins 20 and 21 are both ground. Pin 16 is called "REF WHITE" and is described in the datasheet for the video DAC as follows:
"Reference white control input (TTL compatible). A logical one on this input forces the outputs to a white level, regardless of the input pixel data."
So, since a fully white picture with no relation to the input pixel data is not desired, this input should always be tied to ground in order to keep the DAC from latching into a fully white output state. Therefore, pin 16 should be connected to video ground on the PCB itself, but it isn't on motherboards rev 1A and 1B which is probably a design flaw. In conclusion, this wire is most likely a factory modification which they had to do on all rev 1A and rev 1B motherboards or the Amiga would show nothing but a white picture as you could see in the video here. I have seen this wire on other rev 1A and 1B motherboards as well, but I have never seen it on other revisions as far as I can remember. This also speaks for that this was a design flaw on rev 1A and rev 1B which they fixed in later revisions of the motherboard.
Again, great video as usual. I could use your help on repairing an Amiga 1000 (NTSC) I just received from eBay. Seller said it only "partially" worked. On one TV, all I get is a solid green screen (CRT). On a crappy LCD, I get a solid gray screen. I have re-seated the Agnus and CIA chips. Any ideas?
I haven't really seen any service manual for the A500 on the net, only the schematics (which you could consider being a part of the service manual). But I have seen the service manual for A500+. I don't remember where I downloaded them from, but I remember that I just searched for them on google and found them really quickly.
well on the PLCC each color has 8 pins that represents a byte, thats why for example the B0 switches between high and low more often since its the least significant bit, same goes for green and red.
How come you got a signal from B7 on the oscilloscope when the slider was halfway and upwards but no indication on the probe? It made sense that that the most significant bit was the problem since the intensity reset at exactly halfway, but then the oscilloscope should show a signal? Also can't quite understand why B3 was constantly on from halfway and upwards, expected it to flip twice as much as B4 and half as much as B2?
Awesome!
What, you didn't swap any resistors?? Aw, come on man! You gotta swap the resistors. It's the new trend, you know? And also, there is a big chance of the picture showing up like normal again if you change places of a few resistors at random. :) :)
On a more serious note, I think it might have been something similar to this problem. Was it an A1200 or another model? Since you're writing in past tense, I'm wondering what happened with it? Did you manage to fix it? Or did you decide to sell it?
Your method to create a vertical bar on the screen makes me think of your other screen, with a bar of noise you want to get rid of.. It was kinda noisy instead of the pulses all being perfectly timed, but they did seem to have a cutoff pretty neatly horizontally
Sorry, I'm not really sure what you mean. Which screen is the "other screen"? Is that something that was present in this video, or are you thinking about another video? Could you give me a time mark of where this is seen in the video?
I was sure I watched a vid where you had a display with a vertical bar of noise on the left side and you didn't really have any information available / asked for comments to help, but I can't seem to find it for the life of me.. XD
outseeker
Maybe this video?
watch?v=O-GNE96-yME
Well, it was my 90s Amiga. I was a total noob back then. So the service shop repaired it. Don't know what they did, but it worked. I was too scared to even open my Amiga back then :D A1200 it was. Then I sold it at the turn of the century. Stupidest thing I've ever done. Now I'm overcompensating with multiple Amiga models :D
Maybe I should take one of them now and play "swap the resistors" :) Naaah, I couldn't never do such a thing on purpose :)
And I am so surprised to see you uploaded a new video not after 3 months :D how about the service manual and datasheets? do you know where I can download them for Amiga 500? I still want to upgrade my A500 with bastard motherboard to A500+ if you remember. Currently I am hunting for a dead A500+ on ebaythat hopefully the components I need are still healty in it! but no success yet. Do you have any ?!
Still about? What specific version of Protracker are you using? I've a 1200 with similar problem and all versions of Protracker I've tried so far don't have that colour slider version. Blue is somewhat dim on my 1200 and there's a dark horizontal band when showing fine checkerboard (from Amiga TeskKit)
Interesting... As soon as you showed it was an 8-bit data bus, it was obvious to me that the MSB line was bad. Thinking digitally because of my background. The MSB bit obviously wouldn't turn on.
Btw, I'm curious about this alternative mathematical way to figure out which pin is faulty. I don't know how complicated it is, but if it's possible to explain about this even with the character limitation in the comment section, then please do.
Hi, I have an A600 and the bottom half sceen is darker than the upper half. This problem is on a crt tv, 1084 and even on LCD with a good scart.
About upgrading... Are you sure the layout of your motherboard corresponds with the pinout of an A500+ Agnus-chip? You can not just replace an 8372 Agnus (found on A500) with an 8375 Agnus (found on A500+). The 8372 and 8375 have different pinouts, so you need to have a motherboard that fits the Agnus version you want to use. Search wikipedia for "MOS Technology Agnus", and you will be able the compare the difference in pinout between the different revisions of the chip.
I think it's a good thing to be scared of doing things you don't feel confident with, because that stops people from doing weird things they don't know anythinig about, and really shouldn't be doing. Like for example, swapping resistors at random on motherboards, or replacing surface mounted capacitors on a motherboard with through-hole resistors. :) And also, there was of course this thing about voiding the warranty back in the early 90s that made opening A1200s perhaps not such a smart idea :)
Could a sig gen be used in place of a pulser?
Oh man you Rock
Well the funny thing is, its already has AGNUS 8375 on it :D its just missing the extra memories, audio filter components and the battery and time chip! I will send you a picture in a private message.
Are they using 2x 4-BIT outputs for each colour channel? I figured it was BIT 7, but BIT 3 behaving the same (dropping to 0 at the half-way mark) was a surprise.
Although, if mid-way is where BIT 7 comes on, you'd expect everything to go to 0 at that point - 01111111 -> 10000000.
Actually, I'd expect BIT 3 to go up and down a bit more frequently. I'd expect BIT 6 to come on just before half-way, turn off at half-way and then come on just before full.
Perhaps it was a sampling latency problem, and the slider motion defeated the accuracy.
Another thought:
B0 = BIT 0
B4 = BIT 1
B1 = BIT 2
B5 = BIT 3
B2 = BIT 4
B6 = BIT 5
B3 = BIT 6
B7 = BIT 7
Kind of interleaving the bits across the pins, which would explain the behaviour I observed.
Interesting.
hi retrogamemodz, can you help me with some basic info of what you might think the problem is?
I got a CD32 that i bought from someone as "working"
When i pluged it in, nothing worked, it only showed a green light but it didn't even tried to load a CD.
i open it and saw two bulging caps, 470uf caps if i remember correctly, swaped them and tata, i had sound and the cd was spinning, but i didn't had video, neither from rf or composite, i didn't tried via svideo because i didn't had a cable.
i could see a distorted faint black and white picture when i turned the cd32 on, but i wasn't able to read a thing.
Could this mean that i have a cap issue or could this be a video dac problem or even lisa problem?
Vasco Morais It's difficult to diagnose a problem like that based only on that description. It would help if you could make a youtube video of it and show me what the symptoms look like, and maybe also show me what condition the motherboard is in.
sure, ill try
Cool. Let me know when you have posted the video.
I think the "more beeps (switches)" are because the values are counted in a binary way, so the lower bits should beep more often than the higher beeps. If you count binary, you will recognize this behaviour ;)
why you don't measure resistance between b7 and the ground? it may be short circuit on board under the chip, not broken chip.
Bad conclusion. If such short existed the installation of the replacement chip would have exhibited exactly the previous symptoms.
You can teach whole courses. Wow.
yeah we know someone like that " I know I`ll remove that resistor and put a smoothing capacitor there ! ...............LoL , oh and the music was also excellent ,like I said talent shines through :)
Nicely done! What is with these Amiga owners and their crazy modifications ;)
Hi, your investigations are so cool. Kudos for that!
Here is a problem that's somewhat similar to yours, for which I would appreciate any hint or help:
Amiga 600 Rev 1.5. The Green-Output on RGB shows no output for the first half of the slider, than starts rising. So the Workbench 3.1 background (standard colors) is not grey but purple-ish (missing amount of green). The Composite output ist perfect in colors! So I guess, there is either something wrong with the CXA1145 Encoder or with the components connected to the G_OUT (Pin 22) of the encoder.
(www.amigawiki.org/dnl/schematics/A600_R1.5.pdf / Page 5)
Do you see any other possible suspicious thing?
How can I localize the problem (I have no oscilloscope, but a multimeter)? Any suggestions how to go on?
RetroGameModz, come back soon! Cheers, M
I've got a mobo with those same problems and I would like you to fix it if you place a reasonable quote. I also have got another A1200 mobo that I left for spares and I've got the main components that I took from it (new parts). The thing is that I don't have an idea if you are alive :(
Powered my 1200 on today...she is a bit ruff on the screen..and the sound is nearly non existent...i may have to send to you for repair..i presume it's bad caps
I need you help my amiga 1200 have no pitures, regards from sweden
Dear sir, I would like you to fix my 1200. It has video problems. If you were a surgeon I would gladly put my life in your hands. Genious. Seriously though. Can you fix mine at a price? If so how much?
ta da!
Can I send you my Amiga 600 for repair?
Don't you hate it when Lisa doesn't put out?
me love u
Hi, RetroGameModz, I like your videos, and I decided to ask you about a problem. I've got a really good WD MFM hdd controller, and there's an IC on it, that's pin is connected, and soldered into the pcb, but at the edge of the IC it was broken. It was in the school, so maybe some idiot kids did it. How can I fix it
If you mean that a pin has broken off the IC, then all you can do is to try to get hold of a new IC and replace it on the board.
Ok, thanks. Anyway, I enjoy your videos. :)
Hi there, really nice videos that you have been making, really appreciate it. I have a similar issue with my Amiga 1200, do you happen to know what might be the issue? And could you help me with getting this fixed? (and maybe a new video!) The only mod I have done was apply a 1 uF cap on D215A as I was getting lots of flicker with floppy drive access. Now it's developed this bigger problem with a bright pink screen.
eab.abime.net/showthread.php?p=1042660#post1042660
+tjlazer71
Try changing the slidebars for the three colors (R, G and B) in ProTracker for example to see what you get. I have done this several times in different videos, so you can check that out if you're not sure about what I mean by this.
Check to see if you appear to get a normal response from every one of the three slidebars. That would be the first step when troubleshooting something like that.
+RetroGameModz I will try the Pro Tracker screen color test and see what I get. I've been looking into this problem and I think I have a bad DAC chip, VP101 3BA. The chip is hot to the touch, and when I press on it I sometimes get a screen flash. My board is a 1D1 board and I heard they are prone to this chip going bad. If this is the case, do you know where I can get a VP101 3BA chip from?
+RetroGameModz I did the test with Pro Tracker v3.00. Green is working as it should, and Red and Blue are NOT. So it seems Red and Blue have a bad connection to the DAC.
Yep, or the DAC might be bad. You have to troubleshoot the motherboard with proper test gear in order to find out where the fault actually is. The video DACs can usually be found on eBay if you search for the parts number.
+RetroGameModz Do you offer repair services? :)