What's new

Kujako

Grand Master
Joined
Jun 3, 2020
Messages
1,242
Reaction score
723
Location
USA, California
So as an experiment, I've replaced the battery in a MV1FZ with a 0.47F 5v super-capacitor (I have larger 5F ones, but they're chonky and this little one fits under the cartridge cowl). After leaving it running for awhile, I left it overnight and it's retained the calendar and soft-dip settings. Which would seem to be a positive indicator. But I'm not sure how long it would take to fully "charge", how long that charge would last powered off or even how long it takes for the backup RAM to clear when there's no power given that it retained data during the battery removal and replacement. I had previously tried to run the numbers, but I had to fudge too many of the variables such as voltage delta and amperage draw that it was kind of meaningless (and I don't math gud). So, opening this up for discussion & comments.
1691253379172.jpeg
 
So, one week later after letting the system sit unused it still has the soft dip and calendar settings. Going to call this experiment a success. Based on my incredibly suspect maths it should get around two weeks of shelf time before needing to be reset out of a 0.47F charge.

For those interested, this is what I used.
https://www.digikey.com/en/products/detail/eaton-electronics-division/PM-5R0V474-R/2185064

In a case with more room, I might be able to pull off one of the 5F models. But they're biguns. Would just about fit in a Lyons 1C case. So may give that a try.
1691873253627.jpeg
 
Nice. I will bet you it lasts longer than you think. I've seen stock Turbo Duos hold save data for months and months even though you're supposed to power them on every couple of weeks, even with how old those systems are.
 
Nice. I will bet you it lasts longer than you think. I've seen stock Turbo Duos hold save data for months and months even though you're supposed to power them on every couple of weeks, even with how old those systems are.

Yeah, going to let this one sit I think and try it periodically. As another experiment I went ahead and put one of the 5F monsters in a MV1C I've been working on. The spacing of the pins worked out better than I expected.

https://www.digikey.com/en/products/detail/eaton-electronics-division/PHB-5R0H505-R/2770532

1692145280085.jpeg
 
My one concern about these is the life rating. Will be keeping an eye on them over time. These are listed as 1000 hours at 60 C. Which is low, but that's the point at which capacitance drops not when they fail. And it expands the lower you go temperature wise. Would like to find a chart similar to this for these, but haven't been able to. My understanding, which is probably wrong, is that this is about continuous use under load at temperature and not the actual lifetime of the component.
1692364178664.png
 
Last edited:
nice acrylic case! where did you get/make it ... and the top is backward btw ;)
 
I don't know if this helps/hurts your efforts, but you could use a Higher Voltage Capacitor, as it will only fill up-to whatever the "Charge Voltage" is... as a lot of power supplies have 250v caps as they can then accept europen voltages @ 220 and still work with US/JP Voltages @ 110. So if it's easier to find a higher voltage with a larger MH cap then I'd try that, but typically they're also bigger the higher the capacitance and voltage....
 
I don't know if this helps/hurts your efforts, but you could use a Higher Voltage Capacitor, as it will only fill up-to whatever the "Charge Voltage" is... as a lot of power supplies have 250v caps as they can then accept europen voltages @ 220 and still work with US/JP Voltages @ 110. So if it's easier to find a higher voltage with a larger MH cap then I'd try that, but typically they're also bigger the higher the capacitance and voltage....

Yeah, finding the "right" capacitor has been a challenge. Higher voltage is as you say often larger, but also often have higher ESR which I worry about. Also hard to find one like this with a higher "life" rating, but as I previously said I don't know how much that really matters unless you plan on running things for thousands of hours at a time. I also worry about how long these will last in general. They all say things like "high lifetime cycles" or similar, without actually listing a mean time between failure. Ultimately I don't know if this is a good idea at all... but it was of interest to me and I do see small capacity capacitors being used in consolized systems off aliexpress these days (what looks like around 0.15f capacitors used), so I kind of wanted to see how well those can be expected to work.
 
I mean mtbf is tough to judge on a capacitor, as it depends on temperature, and number of charge / discharge cycles per second / hour / day etc...
 
It's good that you used a 5v cap as the charge circuit on the NiCd is a 470 ohm resistor connected from 5v to limit current.

Check the datasheet for the cap to see if there are charging limits on current that you need to observe to help prevent early failures of the capacitor (shorts internally, heat generation leading to seal failures, etc...)

There are folks that feel it's safe to put a rechargeable lithium battery in place of the NiCd, but a quick look at the data sheet shows how seriously they are overcharging those batteries which can leand to terrible failures with chemicals that hare highly reactive. Luckily, the supercaps won't be as serious, but you don't want to overstress the part.

As for putting the supercap where a rechargeable lithium was is something I'd have to look at the schematic of the circuit for. I haven't seen any MVS schematics for that out in the wild, or even given to me in private, to see what the charge rates would be.
 
It's good that you used a 5v cap as the charge circuit on the NiCd is a 470 ohm resistor connected from 5v to limit current.

Check the datasheet for the cap to see if there are charging limits on current that you need to observe to help prevent early failures of the capacitor (shorts internally, heat generation leading to seal failures, etc...)

There are folks that feel it's safe to put a rechargeable lithium battery in place of the NiCd, but a quick look at the data sheet shows how seriously they are overcharging those batteries which can leand to terrible failures with chemicals that hare highly reactive. Luckily, the supercaps won't be as serious, but you don't want to overstress the part.

As for putting the supercap where a rechargeable lithium was is something I'd have to look at the schematic of the circuit for. I haven't seen any MVS schematics for that out in the wild, or even given to me in private, to see what the charge rates would be.

Wish I could help there, but I got too deep into the weeds just taking measurements and trying to do calculations. What I do know is that I've started to see this (seemingly, not seen it in person only in pictures) in Aliexpress systems. But they're using cheap, low capacity capacitors perfectly positioned to explode all over the RAM. So I wanted to see how well this could actually work.

Example scraped from Aliexpress. These boards are getting more and more modified as time goes on it seems...
1692461356984.png
 
isn't it the case with most batteries like car batteries that you have to increase the voltage to charge them, like 14.5v to charge a 12v battery, so if you have a 5v cap, and it's trying to charge the battery and pumps in 6-7v wouldn't it be better to use a 12v or 8v cap ?
 
isn't it the case with most batteries like car batteries that you have to increase the voltage to charge them, like 14.5v to charge a 12v battery, so if you have a 5v cap, and it's trying to charge the battery and pumps in 6-7v wouldn't it be better to use a 12v or 8v cap ?
Well the voltage is a rating, you can always go higher, but the circuit is providing 5v so that’s a good place to start. Higher voltage rated super capacitors are larger and cost more.
 
Well the voltage is a rating, you can always go higher, but the circuit is providing 5v so that’s a good place to start. Higher voltage rated super capacitors are larger and cost more.
I'm just stating the concern for 5v caps, in response to the question on longevity, as I'm sure there's a 10-20% tollerance built in, and with a charging cuircuit once it feels the voltage is high enough it should stop trying to charge it and just return to monitoring... So it'll get a little jolt of 6-7v in the beginning and then return to normal, but how long before it breaks down due to it pushing the envelope constantly.... (Everytime you turn it on.)
 
I'm just stating the concern for 5v caps, in response to the question on longevity, as I'm sure there's a 10-20% tollerance built in, and with a charging cuircuit once it feels the voltage is high enough it should stop trying to charge it and just return to monitoring... So it'll get a little jolt of 6-7v in the beginning and then return to normal, but how long before it breaks down due to it pushing the envelope constantly.... (Everytime you turn it on.)
In this case, the 5v is a fixed voltage direct feed from the JAMMA power source and from my measurements is actually dropped to around 4.5v by the time it hits the battery connector. Only way for it to be higher than 5v is if your power supply isn't properly adjusted (and then you'll have issues beyond the charging circuit).
 
Back
Top