Pioneer LaserActive PAC-S10 Capacitor Replacement
The 90's were a very interesting time in the world of video games - at least they were through my rose-colored glasses. No one had a clear picture of where the future would take us but it seemed like everyone had a picture. This gave rise to a ton of innovation that companies these days wouldn't dare risk trying.
One such innovation was the Pioneer LaserActive. Basically Pioneer worked with video game companies to create a Laser Disc player with a pluggable modules that contained entire video game consoles. At $1000 for the LD player and $600 per module, it was a contender for the crown of fiscal masochism, easily beating already prohibitively expensive enthusiast-only platforms like the SNK Neo*GEO AES.
Pioneer managed to get both Sega and NEC onboard, so LaserActive offered both a Genesis and Turbografx-16 module. Not only did you get the world's most expensive working Genesis or Turbografx-16 inside your Laser Disc player, plus the functionality of the CD-ROM addons offered for both consoles, you also got access to an exclusive library of games released on Laser Disc. All of the games tried to capitalize on the playback of full motion video (which was really the only reason to use the platform in the first place) and none of them were "killer apps", but they still fetch pretty high prices because most of them are still exclusive to this specific hardware platform.
That's right kiddies, playing LaserActive games on an emulator is pretty much not a thing. This is mostly due to the analog nature of Laser Disc, but more specifically how some LaserActive games used it. Games on Laser Disc aren't exclusive to the LaserActive platform - the original Dragon's Lair arcade machines used LDs, for example. And many if not all of the old arcade LD games can and have been emulated - hell Dragon's Lair can be played on any standard DVD player.
Because LD's aren't digital, they can't be "ripped" in the traditional sense. The material has to be "digitized" (i.e. played back and captured in a digital format like MPEG) in order to be used by an emulator. But with LaserActive it's not a simple matter of digitizing the analog video and re-incorporating it into the program code. In many cases, to speed up scene transformations multiple video streams were recorded on top of one another, taking advantage of the nature of analog interlacing. Scenes were conditionally played back as odd or even scanlines by the LaserActive software. In other words, if you played the raw video into a capture device you'd get an indecipherable visual mess. It's theoretically possible to either "trick" the LaserActive device into playing the properly decoded scenes into a capture device, or building a special capture device capable of decoding the specific method used to encode the video, but all things being equal, it would be extremely difficult to simply "dump" a LaserActive game, and to date I'm unaware of anyone having devoted the time and energy to attempt it.
Because LD's aren't digital, they can't be "ripped" in the traditional sense. The material has to be "digitized" (i.e. played back and captured in a digital format like MPEG) in order to be used by an emulator. But with LaserActive it's not a simple matter of digitizing the analog video and re-incorporating it into the program code. In many cases, to speed up scene transformations multiple video streams were recorded on top of one another, taking advantage of the nature of analog interlacing. Scenes were conditionally played back as odd or even scanlines by the LaserActive software. In other words, if you played the raw video into a capture device you'd get an indecipherable visual mess. It's theoretically possible to either "trick" the LaserActive device into playing the properly decoded scenes into a capture device, or building a special capture device capable of decoding the specific method used to encode the video, but all things being equal, it would be extremely difficult to simply "dump" a LaserActive game, and to date I'm unaware of anyone having devoted the time and energy to attempt it.
A couple of years ago I was able to get one of these players at a pretty reasonable price on Ebay. Mine came with the "PAC-S10" which is the Genesis/Sega CD module. It played "Mega LD" titles, and Sega CD titles just fine, but wouldn't recognize games in the cart slot. While I had never heard of this particular problem, it wasn't hard to guess what it was. The electrolytic capacitors used in electronics in the 90's are hitting an age where their failure rate is spiking. Even the ones that weren't cheaply made or defective have reached the end of their operational lifespan.
Any electrical engineer who would like to chime in and correct me should feel free, but from what I've been able to glean capacitors work sort of like power reservoirs. It's my understanding that this serves a couple of purposes - firstly to "rectify" power - that is keep the amount of power moving steadily between components even when the power given off by the power supply is fluctuating. This is particularly important when dealing with integrated circuits and data processing because the difference between a 1 and a 0 is actually measured by differences in voltage. It may be okay if starting your washing machine in the next room makes your lights dim a fraction, but it would not be okay if your computer started mistaking 1's for 0's because of it! Capacitors determine the amount of power that gets delivered to things like audio and video amplifiers, drive motors etc... When they start to go bad, meaning that they're no longer delivering power at a regular expected level, all sorts of weird things start to happen.
Electrolytic capacitors actually contain liquid electrolyte (yes, what plants crave), which is corrosive and eventually either evaporates (if you're lucky) or leaks out all over the circuit board (if you're not lucky). The failure is based solely on age. New in box or rarely used consoles seem to fail at the same rate as old workhorses. These ticking time bombs are unfortunately super-common because they're cheap and manufacturers don't exactly care if they only last 30 years.
The increasing failure curve of Electrolytic capacitors has brought this problem roaring into the consciousness of classic gaming culture because most of us old guys still gaming on 30+ year old consoles are going to be affected by it sooner rather than later. Now the common remedy to this is to replace all of the electrolytic capacitors in a given console with newer and presumably higher quality electrolytic capacitors - hopefully getting at least another 30 years out of them before they start failing all over again.
When I bought the LaserActive it was with the knowledge that the game modules were notorious for capacitor failure, so I wasn't exactly surprised to find out that my PAC-S10 wasn't completely functional. In fact, alongside it, I acquired a nice soldering rework station against the day I would eventually have to replace the capacitors.
Now, the idea of replacing all of these capacitors with more that would just fail again did not sit well with me. Fortunately not all capacitors are electrolytic. There are quite a few different compositions. Ceramic polymer capacitors were pretty limited and expensive in the 90's but not so much nowadays. But would these really be compatible?
All capacitors are rated for capacitance (imagine that!), voltage and temperature. In this sense one is as good as another irrespective of composition. They do have different properties, however. For example, my quest to determine if ceramic capacitors would be a compatible substitute led me down the rabbit hole of DC Voltage Bias. There's also the fact that most of the ceramic capacitors I found were bipolar - that is to say that polarity does not matter. Almost all of the electrolytic capacitors in the PAC-S10 were polarized.
Firstly, bipolar capacitors are interchangeable with polar capacitors. In fact, you can build a bi-polar capacitor by simply connecting opposing poles (anode and cathode) of two polar capacitors with the same capacitance rating. So there is no problem replacing a polarized capacitor with a bi-polar capacitor.
As for the DC Voltage Bias, that's a trickier subject to tackle. The question is will the DC Voltage Bias property of the ceramic capacitors cause issues if they're used in place of the original electrolytic capacitors. The answer is that it depends greatly on what the capacitor is meant for. For example, audio circuits use capacitors to regulate the frequency of signals based on a constant capacitance value. The higher DC Voltage Bias of a ceramic capacitor causes capacitance to fluctuate when voltage is higher and can actually degrade the quality of the audio signal by altering the signal outside the desired frequency. I'm nowhere near skilled enough to tell which circuits on the board have to do with audio with any degree of reliability, so this is a bit of a problem. The solution seems to be relying on the voltage rating of the capacitor itself. The DC Voltage Bias is expressed as a curve - the variation of capacitance increases exponentially as you reach the capacitor's maximum voltage value. If that maximum value is considerably higher than the voltage is ever likely to get, the frequency variation should remain inconsequential. By selecting a ceramic capacitor of a higher voltage value (say, double) you can minimize fluctuation by keeping the DC Voltage Bias on the low end of the curve.
Using these assumptions I decided to go ahead with the capacitor replacement using ceramic capacitors. I found a list of capacitors on the system and ordered the replacements from Mouser.
Having never worked with surface-mount components before there was a bit of a learning curve. I had never used a hot-air workstation before so I had to browse around YouTube for some pointers and examples. The basic idea is that you blow super-hot air onto the component and the heat passes through and under the component to melt the solder. The fact that components survive being cooked like this was somewhat amazing, but evidently they're all expected to endure that level of heat for short periods of time. After a little experimentation I was able to get an idea of how long I needed to hold heat to a SMD capacitor before it would come loose from the board. Just hold on the component, wave it slightly to make sure the heat is applied evenly to both sides, nudge with the tweezers every couple of seconds until it moves then pluck it off and repeat. After the better part of 100 capacitors I went after the board with a de-soldering braid to clear off any excess solder, then used alcohol and cotton swabs to scrub melted flux and what I assume was leaked electrolyte from the board. There was a similar learning curve with installing the replacement capacitors. By the time I got to the end, I had to re-do the first 10 or 15 to make them look nicer.
Replacing 90-some odd capacitors is a lot of work - especially if you're learning as you go. By the time I put the last one in, my concentration was a little frazzled, but I was really anxious to see the fruits of my labor. I quickly and, I thought, carefully reassembled the PAC-S10, popped it into the LaserActive player, popped a cartridge in and turned it on. I was delighted when the screen lit up and the game actually started playing from the cartridge - that part was fixed. But then I noticed there was no sound and my heart sank as I checked and re-checked all of the audio cables and the receiver. As I sat down to think through what might have gone wrong, I realized I had no memory of re-attaching the daughter board when I reassembled the PAC-S10. I went back to my work area, and sure enough, the daughter board was still sitting there waiting patiently for me to figure out what had gone wrong. With the unit _completely_ re-assembled the sound was back. I played through about 3 stages of Lightning Force, 2 stages of The Terminator on Sega CD, and 2 stages of Hyperion on MegaLD before I was satisfied that everything was working.
A couple of weeks later I was playing around with Laserdiscs and decided to pop in one of the Star Wars Definitive Edition discs and I noticed a low buzzing noise when the sound was turned up. After poking around I noticed that the buzz went away when the Laserdisc was played with the PAC module ejected, so the buzz was definitely coming from it. Problem is I didn't really do any before and after testing so I don't really have any way of knowing if this was a result of using ceramic capacitors, or if it was just poor shielding of the PAC module causing some minor noise. No way I'm replacing all of the capacitors again to find out!
Any electrical engineer who would like to chime in and correct me should feel free, but from what I've been able to glean capacitors work sort of like power reservoirs. It's my understanding that this serves a couple of purposes - firstly to "rectify" power - that is keep the amount of power moving steadily between components even when the power given off by the power supply is fluctuating. This is particularly important when dealing with integrated circuits and data processing because the difference between a 1 and a 0 is actually measured by differences in voltage. It may be okay if starting your washing machine in the next room makes your lights dim a fraction, but it would not be okay if your computer started mistaking 1's for 0's because of it! Capacitors determine the amount of power that gets delivered to things like audio and video amplifiers, drive motors etc... When they start to go bad, meaning that they're no longer delivering power at a regular expected level, all sorts of weird things start to happen.
Electrolytic capacitors actually contain liquid electrolyte (yes, what plants crave), which is corrosive and eventually either evaporates (if you're lucky) or leaks out all over the circuit board (if you're not lucky). The failure is based solely on age. New in box or rarely used consoles seem to fail at the same rate as old workhorses. These ticking time bombs are unfortunately super-common because they're cheap and manufacturers don't exactly care if they only last 30 years.
The increasing failure curve of Electrolytic capacitors has brought this problem roaring into the consciousness of classic gaming culture because most of us old guys still gaming on 30+ year old consoles are going to be affected by it sooner rather than later. Now the common remedy to this is to replace all of the electrolytic capacitors in a given console with newer and presumably higher quality electrolytic capacitors - hopefully getting at least another 30 years out of them before they start failing all over again.
When I bought the LaserActive it was with the knowledge that the game modules were notorious for capacitor failure, so I wasn't exactly surprised to find out that my PAC-S10 wasn't completely functional. In fact, alongside it, I acquired a nice soldering rework station against the day I would eventually have to replace the capacitors.
Now, the idea of replacing all of these capacitors with more that would just fail again did not sit well with me. Fortunately not all capacitors are electrolytic. There are quite a few different compositions. Ceramic polymer capacitors were pretty limited and expensive in the 90's but not so much nowadays. But would these really be compatible?
All capacitors are rated for capacitance (imagine that!), voltage and temperature. In this sense one is as good as another irrespective of composition. They do have different properties, however. For example, my quest to determine if ceramic capacitors would be a compatible substitute led me down the rabbit hole of DC Voltage Bias. There's also the fact that most of the ceramic capacitors I found were bipolar - that is to say that polarity does not matter. Almost all of the electrolytic capacitors in the PAC-S10 were polarized.
Firstly, bipolar capacitors are interchangeable with polar capacitors. In fact, you can build a bi-polar capacitor by simply connecting opposing poles (anode and cathode) of two polar capacitors with the same capacitance rating. So there is no problem replacing a polarized capacitor with a bi-polar capacitor.
As for the DC Voltage Bias, that's a trickier subject to tackle. The question is will the DC Voltage Bias property of the ceramic capacitors cause issues if they're used in place of the original electrolytic capacitors. The answer is that it depends greatly on what the capacitor is meant for. For example, audio circuits use capacitors to regulate the frequency of signals based on a constant capacitance value. The higher DC Voltage Bias of a ceramic capacitor causes capacitance to fluctuate when voltage is higher and can actually degrade the quality of the audio signal by altering the signal outside the desired frequency. I'm nowhere near skilled enough to tell which circuits on the board have to do with audio with any degree of reliability, so this is a bit of a problem. The solution seems to be relying on the voltage rating of the capacitor itself. The DC Voltage Bias is expressed as a curve - the variation of capacitance increases exponentially as you reach the capacitor's maximum voltage value. If that maximum value is considerably higher than the voltage is ever likely to get, the frequency variation should remain inconsequential. By selecting a ceramic capacitor of a higher voltage value (say, double) you can minimize fluctuation by keeping the DC Voltage Bias on the low end of the curve.
Using these assumptions I decided to go ahead with the capacitor replacement using ceramic capacitors. I found a list of capacitors on the system and ordered the replacements from Mouser.
Having never worked with surface-mount components before there was a bit of a learning curve. I had never used a hot-air workstation before so I had to browse around YouTube for some pointers and examples. The basic idea is that you blow super-hot air onto the component and the heat passes through and under the component to melt the solder. The fact that components survive being cooked like this was somewhat amazing, but evidently they're all expected to endure that level of heat for short periods of time. After a little experimentation I was able to get an idea of how long I needed to hold heat to a SMD capacitor before it would come loose from the board. Just hold on the component, wave it slightly to make sure the heat is applied evenly to both sides, nudge with the tweezers every couple of seconds until it moves then pluck it off and repeat. After the better part of 100 capacitors I went after the board with a de-soldering braid to clear off any excess solder, then used alcohol and cotton swabs to scrub melted flux and what I assume was leaked electrolyte from the board. There was a similar learning curve with installing the replacement capacitors. By the time I got to the end, I had to re-do the first 10 or 15 to make them look nicer.
Replacing 90-some odd capacitors is a lot of work - especially if you're learning as you go. By the time I put the last one in, my concentration was a little frazzled, but I was really anxious to see the fruits of my labor. I quickly and, I thought, carefully reassembled the PAC-S10, popped it into the LaserActive player, popped a cartridge in and turned it on. I was delighted when the screen lit up and the game actually started playing from the cartridge - that part was fixed. But then I noticed there was no sound and my heart sank as I checked and re-checked all of the audio cables and the receiver. As I sat down to think through what might have gone wrong, I realized I had no memory of re-attaching the daughter board when I reassembled the PAC-S10. I went back to my work area, and sure enough, the daughter board was still sitting there waiting patiently for me to figure out what had gone wrong. With the unit _completely_ re-assembled the sound was back. I played through about 3 stages of Lightning Force, 2 stages of The Terminator on Sega CD, and 2 stages of Hyperion on MegaLD before I was satisfied that everything was working.
A couple of weeks later I was playing around with Laserdiscs and decided to pop in one of the Star Wars Definitive Edition discs and I noticed a low buzzing noise when the sound was turned up. After poking around I noticed that the buzz went away when the Laserdisc was played with the PAC module ejected, so the buzz was definitely coming from it. Problem is I didn't really do any before and after testing so I don't really have any way of knowing if this was a result of using ceramic capacitors, or if it was just poor shielding of the PAC module causing some minor noise. No way I'm replacing all of the capacitors again to find out!
Thanks for the write-up - I'm looking to use ceramics for a re-cap of a Sega PAC as well and wonder if you ever solved the buzzing issue when playing LD's or noticed any other issues that could be related to using ceramic caps. Also, did you end up buying all double voltage value capacitors?
ReplyDeleteSince I only have one PAC-S10 there are only two real options for seeing if the noise is capacitor related or something else - either get another PAC module, or swap capacitors. Haven't had the money to do the former or the time to do the latter. As I mentioned earlier the noise is quite subtle so I had completely forgotten about it before you asked.
ReplyDeleteIIRC I wasn't able to find everything in double value, but what I could, I did. The biggest issue is availability. Mouser, Digikey and other suppliers seem to run out of certain components and won't restock them unless there's a giant order from somewhere. I spent a whole day researching part numbers from 3 or 4 different suppliers only to find out that when I went to place my orders two weeks later, some of them were indefinitely out of stock. My advice is be patient.
Thanks for your response! I was browsing through a PAC supplemental service manual and found an issue that sounded like the one you described regarding the buzzing. Apparently there was a service bulletin and the fix was replacing ICs in both the CLD-A100 and Sega PAC that would mute FM output from the PAC to the player when playing laserdiscs. There's some other interesting bulletin bits in there as well if you are so inclined:
ReplyDeletehttps://console5.com/wiki/File:Pioneer_CLD-A100_Supplemental_Service_Information.pdf
Thanks for the tip. That definitely explains what's going on. Although it seems like there may be some reprogramming involved to fix it. I didn't notice it until I read the advisory, but I definitely don't hear the noise when playing games - only when playing laserdiscs. I'll post an update if I ever feel adventurous enough try the fix.
ReplyDeleteWould be awesome if you had a list of those caps. I'm about to redo my unit and ceramic caps would be nice
ReplyDelete