Colecovision internal electronics maintenance after 30 years

[Dennis Stith]

As it goes, people still use Apple ][ computers as well as colecovision systems. I have found over the years that, depending on how much use a system has had, some components dissepate and degrade over time.

 

The primary components that can cause trouble in electronics usually, for me and anyone can post in this thread if they have found any other things, are capacitors. The Colecovision set has several capacitors in it. The tall round ones are what are called "electrolytic capacitors" which basically are comprised of a small aluminum barrel, a piece of material that has been dipped in electrolyte acid, and a couple of leads coming out which are then soldered to the main circuit board.

 

Now, why do I bring this up here? Well I have had to replace capacitors on electronics which are much newer than the Colecovision before and often times could see that some were beginning to bulge on those devices. The bulge means that the electrolyte dissepated and formed a vapor inside the capacitor and is trying to get out. If they burst, that is when the vapor actually was able to leave its housing. The Coleco set I have does not have any bulged capacitors, however since they are so old the electrolyte could have made it out a different way.

 

I have taken physics and DC circuits courses in college and have found that capacitors are not always the easiest things to test and especially if they are still soldered into a circuit. Generally speaking, it is easier to remove capacitors that you think are bad and replace them than to go about trying to figure out how to test them on the board or remove then test them. This is mainly due to the fact that such small capacitors in limited quantities are rather inexpensive. For example, I spent less than $10 replacing all the capacitors on a circuit board in a Samsung 21" widescreen computer monitor whereas if the circuit board was available through OEM, it would have cost $45 or more. And to cross relate this to a colecovision, the pricing may be somewhat similar as to purchasing capacitors and replacing the entire system with another used one.

 

What I am wondering is if anyone knows what the purpose of the capacitors on the main PCB of this are? I am assuming that the beish ones may be what is referred to as "filter capacitors" which does just that, they filter stray signals out to clean up the direct current that is flowing through the computer. Now, the barrel type electrolytic capacitors on the other hand can be used for this as well and also to temporarily store electricity. If for some reason all of the capacitors on this have dissepated and are not performing any more, and the set still works, to me that means that although they are there possibly to filter current, they might not really be entirely necessary. Basically I like to keep electronics the way that they were intended to be which is why I am wondering if I should purchase all new caps for this system and replace them all.

 

Also I'm not sure if there are other components on this that can suffer from age. If anyone knows, then it can be useful to some people that are trying to restore older Coleco sets.

 

In computers, I have found that modern systems only last less than 10 years due to the amount of use they get. Colecovisions were not usually run 40 hours a week like a business workstation. They are older though and components tend to have a lifespan irregardless of use. My Coleco still works and I'm sure that you can go by "if it isn't broken, don't fix it" but if it is good practice to replace anything, then feel free to post your experiences in this thread.

 

I'm hoping that any of my ideas on this forum can help out more people that are Coleco hobbyists and thanks for the help I have received on the other threads. I'm just going to replace the RF jack and then possibly do the RCA jack mod afterwards all the while making sure that I get proper flow while soldering. I'm normally used to soldering inside of newer systems with much smaller wires and connection points so this Coleco should be very easy to work with.

[cybercylon]

Regarding why capacitors tend to fail sooner in more recent hardware, you may want to check this out:

 

http://en.wikipedia.org/wiki/Capacitor_plague

[Dennis Stith]

yeah that details electrolytic capacitors. I saw a few on the coleco board. but there are also ceramic capacitors which are light brown disc shaped solid state components. I'm not sure if those go bad but it looks like those are made from alternating layers of metal and ceramic only so probably nothing to worry about.

 

there are so few electrolytic capacitors on this, it might be a good idea to replace them since the system is so old. they probably aren't doing anything any more anyway.

 

Regarding why capacitors tend to fail sooner in more recent hardware, you may want to check this out:

 

http://en.wikipedia....apacitor_plague

[fiddlepaddle]

Being a software guy who never really understood why electrical engineers "threw a bunch of capacitors" on a board design, often with no specific justification, only a general "they're needed" upon inquiry: this is an interesting thread for me.

 

I know some equipment continues to operate after obvious cap failure, and adding an analog component to a digital design never felt like the right solution anyway, although I understand that the "digital" part of the design is actually just a shorthand method to describe what is actually going on.

 

The other interesting thing is why in the world do we have to put components containing liquids on a circuit board, on equipment that could be (theoretically) designed to last decades, if not centuries? It seems like storage of small amounts of power could be done with solid components...

[Dennis Stith]

They don't just store power, they also filter it so it gives a clean signal. My first project was CCC (Carbon Copy Card) for Gameboy which is how I may have first come to understand filter caps. In its schematic, the author of the PCB stated where to put most of the capacitors and he gave a diagram to follow. The circuit board did not have holes for all of the capacitors in his diagram and so the end user is instructed to put them there only if he wants to because they are filter caps. All of the filter caps on that circuit were the ceramic type. I remember it has one large electrolytic capacitor on it (which I once mistakenly thought was bad even though it wasn't bulged). I replaced it once but later found out it was the power switch that was bad. Since the power switch was double pole, I just desoldered it, turned it around, and soldered it back in. Works.

 

I'm uncertain if the ceramic type have electrolyte in them but it is doubtful so far. So I don't think all capacitors have "liquid" inside of them. But you have to understand that this liquid is soaked up in a paper/sponge like material so it isn't really ever going to leak out onto your PCB anyway. Also the electrolyte only dissepates in small quantities into the air and so everything is going to stay relatively dry unless maybe you toss the coleco into your bath tub and turn on the faucet

 

Storage of small amounts of power can be done with solid components. There are "solid capacitors" now which most often are surface mount and still contain electrolyte. This electrolyte is mixed within a polymer, rather than being soaked in a sponge, and the whole thing basically is a solid prior to and after assembling. If you have a PC built within the past few years, take a look at the main board and you will see them on it rather than the usual electrolytic capacitors that are commonly seen on Pentium 4 and early Core2Duo motherboards. These solid capacitors take the place of electrolytic ones and also have a longer service life. I haven't attempted to replace an electrolytic capacitor in a circuit with a solid capacitor, but since electrolytic capacitors are usually so easy to replace I haven't really had a need to yet.

 

Long ago, I found that if you calculate it, you have to keep a PC on 24 hours a day, all year for 3 years straight for the capacitors on the motherboard to go bad. There are some motherboards that continue to function longer and this probably means that the capacitors were mostly used to filter current and do not break the operation of the system when their life span has been exceeded. However, of course, there have been motherboards that failed prematurely, such as ASUS Pentium 4 boards primarily P4P800 series models, and that was due to the manufacturer choosing an inferior quality capacitor/manufacturer for it systems boards. I have a Pentium 4 P4P800 which is currently on its third P4P800 motherboard more than likely due to this. I'm not even sure why that computer still works today.

 

Anyway it is possible that an engineer might tell you "they're needed" just to make things simple because they are used to a lot of people in their day to day life that don't want to listen to detail on why they do things a certain way (e.g. they won't understand anyway so let's give them a simple answer) However the ones I come across online always give me the proper explanation and in case of the project that I soldered together years ago, the CCC, it was clearly stated that they were "filter caps" (the ceramic ones anyway).

 

Being a software guy who never really understood why electrical engineers "threw a bunch of capacitors" on a board design, often with no specific justification, only a general "they're needed" upon inquiry: this is an interesting thread for me.

 

I know some equipment continues to operate after obvious cap failure, and adding an analog component to a digital design never felt like the right solution anyway, although I understand that the "digital" part of the design is actually just a shorthand method to describe what is actually going on.

 

The other interesting thing is why in the world do we have to put components containing liquids on a circuit board, on equipment that could be (theoretically) designed to last decades, if not centuries? It seems like storage of small amounts of power could be done with solid components...

[SlowCoder]

Very interesting, and informative thread. Thanks!

[Dennis Stith]

Alright so I finally got some time to take this back apart and take a look inside. I see a whole bunch of ceramic capacitors. I'm not sure how long these last so I'm going to leave them alone. They appear to be some sort of dry capacitor technology.

 

The electrolytic capacitors inside the colecovision set consist of the following: one 50V 1.0uF, four 16V 10uF, and one 50V 10uF capacitor (on the RF board).

 

All of these capacitors haven't blown but I assume due to how old this set is, they probably have completely dissepated by now and possibly are all open circuit anyway. It is late now so I'm not going to test the leads and see if they still work. I intend to replace all six of these capacitors with brand new ones.

 

There is one more place that has capacitors though. I came across a video on youtube of a guy who dremel cut the power supply open and found one bad capacitor inside there. The power supply itself has some larger capacitors in it and unfortunately the only way to open it is to cut it open. I want to find a way to re-seal the power supply back together before I go inside of one.

 

I own two coleco vision sets so one of them I'm going to fully go over and replace some parts including the DRAM and the other one I'm probably going to leave alone for the time being. Although I probably should replace the RF to composite modification in both units with the LM318N circuit board and see if I can find out what needs to be done to the audio circuit as I was reading last night that directly wiring from the texas instruments IC to the composite out for mono is a bad idea.

 

I'll try to come back and post on here soon with part numbers from digikey so people know what to buy.

 

Keep in mind that this maintenance I'm doing does not appear to be necessary presently since I haven't actually tested the capacitance of the parts that are still on this board. This is just for information only.

[+5-11under]

There's plenty of items in the CV that will die before the capacitors. I'm not sure why you're hung up on them.

Anyway, for the composite video mod, here's a simple one. In the least it shows where to grab audio from:

http://www.atariage.com/forums/topic/163446-diy-modifications-for-colecovision/page__view__findpost__p__2019142

[SlowCoder]

There's plenty of items in the CV that will die before the capacitors. I'm not sure why you're hung up on them. Anyway, for the composite video mod, here's a simple one. In the least it shows where to grab audio from: http://www.atariage.com/forums/topic/163446-diy-modifications-for-colecovision/page__view__findpost__p__2019142

Your statement may very well be right. And it probably is. Especially true for the ESD sensitive components.

 

But caps are a regular component to check when something goes awry, in any circuit, so I wouldn't be so hard on him for concentrating on them.

[opcode]

There's plenty of items in the CV that will die before the capacitors.

 

Hopefully not the users...

[+5-11under]

There's plenty of items in the CV that will die before the capacitors.

 

Hopefully not the users...

I guess some of us are really "in"to our CVs, then.

[Bruce Tomlin]

Why are you so fixated on electrolytic capacitors? I've never had one be a problem in a CV. In old stuff, they're usually only a problem with the high-voltage ones in coin-op displays.

 

Here's my list with the CV:

 

1) The power switch. They used inferior grease that (even ten years ago) was all caked up into crap, leaving the switch mechanism to oxidize. Disassemble the switch, clean, and re-grease with dielectric grease from an auto parts store.

 

2) The controller port input chips. They may be LS TTL, which is pretty resistant to static electricity, but they do go out. I suspect it is due to the design of the port being able to put -9 volts on one of the pins, and some mechanism by which it gets shorted to an input pin. TTL may be resistant to static electricity, but it is not resistant to voltages outside the power supply range.

 

3) The video DRAM chips. These tend to go bad, possibly due to bad power from problem #1 above. (4116 and earlier chips needed +12 and -5 volt power, and thus would be more sensitive to power problems.) This causes "scrambled" graphics, usually with the startup screen displaying the wrong text characters, which can be used to identify which chip is causing the problem.

 

Beyond that, just the usual problems with connections failing.

[+Yurkie]

To add to Bruce's list.

 

4. VDP, it is almost as likely perhaps even more likely to go bad than a DRAM

5. The RF module

 

Biggest thing to keep in mind is like the old slogan goes "If it ain't broke, don't fix it"

[chr1s]

 

 

Here's my list with the CV:

 

1) The power switch. They used inferior grease that (even ten years ago) was all caked up into crap, leaving the switch mechanism to oxidize. Disassemble the switch, clean, and re-grease with dielectric grease from an auto parts store.

 

2) The controller port input chips. They may be LS TTL, which is pretty resistant to static electricity, but they do go out. I suspect it is due to the design of the port being able to put -9 volts on one of the pins, and some mechanism by which it gets shorted to an input pin. TTL may be resistant to static electricity, but it is not resistant to voltages outside the power supply range.

 

3) The video DRAM chips. These tend to go bad, possibly due to bad power from problem #1 above. (4116 and earlier chips needed +12 and -5 volt power, and thus would be more sensitive to power problems.) This causes "scrambled" graphics, usually with the startup screen displaying the wrong text characters, which can be used to identify which chip is causing the problem.

 

Beyond that, just the usual problems with connections failing.

 

I've got a CV that's now displaying "scrambled" graphics at startup, after having powered up just fine just a week ago. The CV is fully functional with a 2600 adapter, so it seems these DRAMs could be the culprit? All three p/s voltages are getting to the main board, and I've disassembled, cleaned and lubricated the on/off switch. How could I go about identifying a faulty DRAM?

Edited April 9, 2012 by chr1s

[Dennis Stith]

I did the power switch repair but since I put in LM318N now, it appears the problem is back probably because that LM318N takes power directly from this finnicky switch. But it should work. I posted a new thread about it anyway. The controllers don't have any issues unless there is something besides playing the games that I should be noticing with them. I replaced all 8kb DRAM with the +5V mod and the new ones work fine.

 

Why are you so fixated on electrolytic capacitors? I've never had one be a problem in a CV. In old stuff, they're usually only a problem with the high-voltage ones in coin-op displays.

 

Here's my list with the CV:

 

1) The power switch. They used inferior grease that (even ten years ago) was all caked up into crap, leaving the switch mechanism to oxidize. Disassemble the switch, clean, and re-grease with dielectric grease from an auto parts store.

 

2) The controller port input chips. They may be LS TTL, which is pretty resistant to static electricity, but they do go out. I suspect it is due to the design of the port being able to put -9 volts on one of the pins, and some mechanism by which it gets shorted to an input pin. TTL may be resistant to static electricity, but it is not resistant to voltages outside the power supply range.

 

3) The video DRAM chips. These tend to go bad, possibly due to bad power from problem #1 above. (4116 and earlier chips needed +12 and -5 volt power, and thus would be more sensitive to power problems.) This causes "scrambled" graphics, usually with the startup screen displaying the wrong text characters, which can be used to identify which chip is causing the problem.

 

Beyond that, just the usual problems with connections failing.

 

I'm not sure what VDP is but I will definitely look into that. I just looked it up, TMS9928A Texas Instruments Video Display Processor. I don't see any sites listing a replacement part for it though mine appears to work. The RF module works fine but I haven't replaced its capacitor yet (yeah I know it doesn't need to be replaced but I have a brand new one so if I get around to it I might as well. it actually doesn't hurt anything, sans the "if it isn't broke don't fix it". You can always do maintenance on things if you know what you are doing and you won't break it. However if someone doesn't know how to work on the coleco without burning the pcb, breaking traces, or burning up solid state components then good luck )

 

To add to Bruce's list.

 

4. VDP, it is almost as likely perhaps even more likely to go bad than a DRAM

5. The RF module

 

Biggest thing to keep in mind is like the old slogan goes "If it ain't broke, don't fix it"

Edited April 22, 2012 by Dennis Stith