What's new

Hatsune Mike

Champion
Joined
Dec 8, 2017
Messages
736
Reaction score
2,122
Location
Tokyo, Japan
I've created a Digikey BOM for a special MS9 cap kit. Wherever possible, solid-state capacitors (Aluminum Poly, Ceramic, etc) are used in place of electrolytic caps. This covers all of the signal section, with the intent of guaranteeing more linear response curves for the RGB inputs, and long-lasting stable geometry that isn't as temperature sensitive. Because of limitations of capacitor technology, much of the primary power supply section remains in the form of electrolytic caps. All are rated for 105 degree temperature, and are based on the BOM of my normal MS9 electrolytic cap kit.

Here's a Digikey cart: https://www.digikey.com/short/zqwq20

BOM attached in a ZIP file.

ogTN2HX.jpg


The downside is that it's quite expensive (about $55 USD) but I hope the (ostensible) long term stability offsets the cost in the long run.

I replaced the caps in my MS9 this way the other night, and ran it for about 4-5 hours. I could detect no adverse effects on my monitor's image.

Capacitors have reference designators defined, so each cap should arrive in a nice clear baggie with the capacitor location written on it.

Please excuse my poor tube geometry and convergence; adjusting the yoke and doing convergence strips is a royal pain in the Egret II since the monitor is frameless.

C1absuu.jpg
 

Attachments

  • ms9-highlife.zip
    1.4 KB · Views: 218
Last edited:
i dont have a scematic to see what you did,
but i hope you understand the properties of different cap technologies.
 
I just want to say that I strongly support this. A lot of capacitor replacement websites and kits use cheapo stuff.

Electrolytics are known for their poor characteristics and short life. The only reason we see them everywhere is because they're cheap to make and usually "good enough."

Just this week I repaired a less than 2 year old HD monitor because of crappy electrolytic capacitors.

Aluminum polymer is the way to go for stuff like this. It's a newer technology and vastly improved. Of course, the drawback is cost. But when were talking about irreplaceable vintage Japanese monitors (that are gorgeous, I might add) I think it's well worth it :)

Thanks for sharing!
 
I made a MS8-26S equivalent of this cap kit. Works well, be careful while installing because some of the caps have conductive housings and are very close to jumpers. https://www.digikey.com/short/zp30h0

It only has a couple of minor improvements needed - it's missing C560, and C572, C913, C909, and C437 could use wider lead spacing. I haven't included these changes because I haven't tested them.
 

Attachments

  • ms8cap.txt
    3.5 KB · Views: 198
Electrolytics are known for their poor characteristics and short life.
Short life is relative, the real factor is manufacturing quality.

I recapped both PSUs in my cab, one was a MeanWell the other a Eago (super cheap).
First, no capacitors had leaked or bulged on the MeanWell whereas half of them where in poor shape on the Eago. Then out of curiosity I checked ESR and they were all spot on on the MeanWell but NONE was good on the Eago. My cab is from 1991, that's almost 30 yo capacitors, still I'm confident the ones in the MeanWell would have lasted a couple decades more.

That said, a longer lasting solution, even at a higher cost (but looks reasonable to me), is of course the way to go.
 
so just to be clear, polymers DO age.
i have tested and binned lots of old polymers from server motherboards.
 
Where they really shine is in lower duty cycle and lower temperature applications. At 105C they are usually only rated at 3000 hours, whereas you can find high life liquid electrolytics with up to 12000 hours of life at 105C. But they have a different temperature derating curve:

Table_1_Lifetime_overview_with_different_ambient_temperatures.jpg
 
I've created a Digikey BOM for a special MS9 cap kit. Wherever possible, solid-state capacitors (Aluminum Poly, Ceramic, etc) are used in place of electrolytic caps. This covers all of the signal section, with the intent of guaranteeing more linear response curves for the RGB inputs, and long-lasting stable geometry that isn't as temperature sensitive. Because of limitations of capacitor technology, much of the primary power supply section remains in the form of electrolytic caps. All are rated for 105 degree temperature, and are based on the BOM of my normal MS9 electrolytic cap kit.

Here's a Digikey cart: https://www.digikey.com/short/zqwq20

BOM attached in a ZIP file.

ogTN2HX.jpg


The downside is that it's quite expensive (about $55 USD) but I hope the (ostensible) long term stability offsets the cost in the long run.

I replaced the caps in my MS9 this way the other night, and ran it for about 4-5 hours. I could detect no adverse effects on my monitor's image.

Capacitors have reference designators defined, so each cap should arrive in a nice clear baggie with the capacitor location written on it.

Please excuse my poor tube geometry and convergence; adjusting the yoke and doing convergence strips is a royal pain in the Egret II since the monitor is frameless.

C1absuu.jpg
"Time for combat. Good luck!" I'm jealous to see Thunder Dragon 2 in your collection running on an MS9.
 
I made a MS8-26S equivalent of this cap kit. Works well, be careful while installing because some of the caps have conductive housings and are very close to jumpers. https://www.digikey.com/short/zp30h0

It only has a couple of minor improvements needed - it's missing C560, and C572, C913, C909, and C437 could use wider lead spacing. I haven't included these changes because I haven't tested them.
Awesome, thanks!!


Electrolytics are known for their poor characteristics and short life.
Short life is relative, the real factor is manufacturing quality.
I recapped both PSUs in my cab, one was a MeanWell the other a Eago (super cheap).
First, no capacitors had leaked or bulged on the MeanWell whereas half of them where in poor shape on the Eago. Then out of curiosity I checked ESR and they were all spot on on the MeanWell but NONE was good on the Eago. My cab is from 1991, that's almost 30 yo capacitors, still I'm confident the ones in the MeanWell would have lasted a couple decades more.

That said, a longer lasting solution, even at a higher cost (but looks reasonable to me), is of course the way to go.
Its not exactly relative. I say short (well, shorter) life because electrolytics are more prone to drying out after years in a hot cab. That said, brand makes a *huge* difference. My friend has a Amiga CRT from the mid 80s with regular caps that still works great :)
 
@Segasonicfan I have loads of boards from the 70s and 80s with perfectly fine electrolytic capacitors. Might be shorter but still very impressive (40 to 50 yo). Also means if they fail tomorrow and I replace them with as good parts (if that still exists...) then I'll be good for my time on earth.
 
@Segasonicfan I have loads of boards from the 70s and 80s with perfectly fine electrolytic capacitors. Might be shorter but still very impressive (40 to 50 yo). Also means if they fail tomorrow and I replace them with as good parts (if that still exists...) then I'll be good for my time on earth.
Yeah...it just depends. Where and how they are used in the circuit, tolerances, manufacturing quality, etc.
Cause if you look at 20-year-old gaming consoles right now, a lot are failing from capacitor problems.
Electronics from the 70s / 80s are necessarily less prone to this because they are simply less complex. Then you take a look at late 80's and a top of the line Sharp Japanese computer like the X68K which has cap failures everywhere...

The point? Using Alum polymer / ceramics is better, If you can afford it.
 
Here is a good video on going solid state with oscilloscope readings.


and a few supplemental articles/knowledge on them.

https://www.we-online.com/web/en/el...bcm/blog_detail-worldofelectronics_105536.php

https://electronics.stackexchange.c...all-electrolytic-capacitors-with-ceramic-ones
The much lower ESR of ceramic capacitors (vs. electroytic caps) has a feedback loop stability implication.

https://product.tdk.com/info/en/products/capacitor/ceramic/mlcc/technote/solution/mlcc03/index.html
Polymer can potentially cause oscillations in PSU circuits. Polymer also have lower ESR

From my understanding/reading I wouldn't just replace Electrolytic capacitors with ceramic as a general rule due to their esr. Gametech did this with the N64 PSU replacing all the electrolytic capacitors with ceramic in a n64 psu. Whether this will "future proof" it or not remains to be seen.

One thing to keep in mind is sometimes certain capacitors are chosen for reasons unknown. Sometimes it is what they had on hand, reduce cost, inexperience in design, or did not have a time machine for polymers. Like with all things different things have varying degrees of tolerance.A good example related to this is replacing general purpose capacitors with low esr. 99% of the time you will be fine but there may be a corner case that gets you in the balls.

In any case nice capacitor kit Mike. I hope to see more of these!
 
Last edited:
the tolerance on most ceramics is junk,
i try to replace ceramics with polypropylene in anything other than power smoothing circuits.
 
Ceramics all have tolerance ratings of course. I only use X7R or better.

Theres a lot of considerations with caps in a design. But if you get aluminum polymer with a good brand and the same specs it's almost always going to be leaps and bounds better than cheapo standard electrolytics.
 
Last edited:
Back
Top