A Modern Battery Pack for the Epson HX-20

[tezza]

I'm looking at trying to replace the old NiCAD batteries in my Epson HX-20 with modern NiMH equivalents.

The simplest way would be to simply wire a new battery pack up to the old connector pins. Trouble is a 4 x AA battery pack is just a little too deep for the battery receptacle. It doesn't fit nicely into it.

I wonder if AAA batteries can be used. The battery holder is smaller and it would fit. Would they supply the current necessary though?

Any ideas or experience?

Tez

 

[Erik]

It's been several years now, but I was able to purchase a new HX-20 battery on the net for a reasonable price when I needed one.

 

[nige the hippy]

I've usually used tagged NiMH cells, sticky tape & stiff plastic sheet to make up packs. You can get loads of cell size variants from people like RS (radiospares). I would imagine though that using AAA cells in place of old larger cells would yield good results, the series resistance of nicads & NiMHs is low whatever type you use, and modern cells have hugely increased power densities.

On the subject of batteries I'm looking to replace a couple of AAA size wire ended 3.6V 45mAH nicads in my old gridpad, it's going now after RTC surgery (DALLAS as usual ;-)) the RTC holds the scaling factors for the pen, so no RTC = No input. Anyone know an equivalent?

 

[NathanAllan]

When I needed one, I took the old one apart and used the cells from an old drill battery as a direct replacement. I had to use some tape and no holder was available, but it worked. That was over a year ago and they still work great.

 

[Terry Yager]

tez,

Around here, RadShack sells sub-C cells with tabs for about ten buck$ a pair. You can find them on the Net for even less. They are NiCad, but I would think that @~20 hours per charge, that shouldn't be a problem.
OTOH, I read somewhere that someone (was it Nathan?) did replace theirs with AAs and didn't lose much runtime, so I s'poze it's a viable swap.

--T

 

[Sharkonwheels]

Just be careful, and don't leave it charging out of eyesight.
NiMH and NiCD have different properties, different charging methods, and different methods for peak detection (detecting when it's full).

Using standard NiCD charging methods for NiMH will usually result in shortening the life of the NiMH cells by a factor of 10, as well as possibly overcharging them and causing them to leak, never mind the fact that your computer will be ruined.

I don't know at what charging rate the Epson onboard charger charges at, so I can't really say what will happen. Here's rules of thumb:

NiCD's should be charged at double their rate, ie, a 1000mah NiCD, charge at 2000mah

NiMH's should be charged at 1/2 their rate, ie, a 1000mah NiMH charge at 500mah. Exceeding this will shorten their life dramatically - trust me, it's true. Ruined many a pack, because I needed it recharged faster, but I knew the consequences!

LiPO and LiIon have their own settings, known as "C", so a 1.0C means charge it at 1x it's capacity. I think LiIon and LiPo are general 1.0C charging, meaning if ya had a 5000mah pack, it would take a LOOONG time to charge! "C" is also used to refer to it's discharge rate, so 10C output means 10x it's capacity. for a 2000mah LiPo, for example, the above would be charge at 2000mah, and has a discharge rate of 20000mah (20A). It is NOT unheard of to find LiPo/LiIon cells capable of 20C, 40C, even over 70C discharge rates! see A123racing for some hybrid Lithium cells capable of I think 70C!

Gleaned all this from radio control cars, but it 100% pertains here - batteries is batteries.

Another thing you need to be cautious of, is how you STORE the NiMH, as storing them with full or empty charge can cause them to go bad. LiPo's and LiIons should be stored, I believe, with a 70%charge.

Google "Battery university" for more detailed info. That site has awesome info on batteries.


T

 

[NathanAllan]

tez,

Around here, RadShack sells sub-C cells with tabs for about ten buck$ a pair. You can find them on the Net for even less. They are NiCad, but I would think that @~20 hours per charge, that shouldn't be a problem.
OTOH, I read somewhere that someone (was it Nathan?) did replace theirs with AAs and didn't lose much runtime, so I s'poze it's a viable swap.

--T

Not me, I used sub-C's. I couldn't get the AA's to fit in the machine in any configuration.

By the way, is the serial port on the HX-20 a normal one? I mean the pin assignments, not the port itslef. I tried building an adapter cable once, thinking of doing it again.

Nathan

 

[MikeS]

<snip>Here's rules of thumb:
NiCD's should be charged at double their rate, ie, a 1000mah NiCD, charge at 2000mah
NiMH's should be charged at 1/2 their rate, ie, a 1000mah NiMH charge at 500mah. Exceeding this will shorten their life dramatically - trust me, it's true. Ruined many a pack, because I needed it recharged faster, but I knew the consequences! <snip>
T

----------
???
(milli)amphours are a measure of capacity; don't know what you mean when you talk about charging *rates* of x mah? 2A for an hour? 1ma for 2000 hours? As you say, it's more about current and temperature and not so much about time.

There's an awful lot of misinformation and urban myths out there about batteries; let's not confuse the issue even more...

m

 

[Sharkonwheels]

----------
???
(milli)amphours are a measure of capacity; don't know what you mean when you talk about charging *rates* of x mah? 2A for an hour? 1ma for 2000 hours? As you say, it's more about current and temperature and not so much about time.

There's an awful lot of misinformation and urban myths out there about batteries; let's not confuse the issue even more...

m

By charge rates, I mean if you have a 1000mah NiCD, it's max charge rate will usually be 2000ma (2a) for 30min.
You're nitpicking...
The problem will not be, in this case, of the charging rate, but peak detection, and the charging circuit knowing when to stop, and go to trickle-charge mode.

Basically, if you exceed a batteries max charging rate you can damage it.
In the radio control workd, we have more precise control over charging, ie, my chargers are all digital, computerized, and support the different peak detection techniques required for charging NiCD, NiMH, LiPo, and LiIon. Also, my BETTER chargers, even have the temp sensor for yet another input to the charging profile/method. And, some of these things can get VERY expensive VERY fast. But, when you pay $50-100 per 6-pack, and have to have 4,5, or even 6 or more packs, the investment in a good charger pays off.

Charging rates are nothing new. If you put in a set of NiMH in a computer that had NiCD's, and exceed the charging rate *AND* time of the NiMH, they wil go bye-bye in an ugly way - usually by first "sweating," to later leaking the acid.

These are NOT urban myths, but true - they have been proven time and time again. You can go to basically any rechargeable battery manufacturer's website, and gleen the above basic info.

And it IS about time, as welll as current temperature. Time would equal overcharging, because NiMH and NiCD use different peak detection methods, the onboard charger would not know when to stop, so it would try to keep shoving into the cells at max charge rate. Or, it may stop early, and not get near full charge, which isn;t a problem, you'll just get less life from the cells because they weren't fully charged.

Even though one major myth is that NiMH do not have memories, that is not entirely true. I routinely have to "cycle" my packs to get them back to max capacity.

You can basically go to an RC site, find a manufacturer of chargers (Integy, Duratrax, Futaba, MRC, etc..) and download their manual, and pretty much all of them describe what I have attempted to describe. There IS a difference in cell chemistries, and they DO require different charging methods.

I have cases of sub-c's in the garage, and if it were that easy, I woulda' changed my PX-8 over to NiMH years ago.
But I know better.
I also would've done the NEC PC-8201a's, but instead, I used NiCD's in the NEC, and the PX-8 pack is still working up to now.

T

 

[Sharkonwheels]

As an example, the below is the instruction for the Duratrax Piranha Digital peak Charger (I own one) - the bold print is their bold print, not me pointing out anything:

Follow these steps to configure a memory setting:
1. While the charger shows the default display, press the (+) button to access the “BATT SETUP” menu.

2. Briefly press ENTER. The “BAT.TYPE” setting will be displayed. Press the (+) or (-) buttons to set either “NiCd” or “NiMH” to match the exact chemistry of the battery to be charged. This tells the charger exactly how to identify peak charge for the battery to be charged. Press ENTER to confirm the setting.

3. For “CAPACITY,” press (+) or (-) to set the charge input capacity to the desired mAh (milli-amp hour) level. This selection ranges from 50mA to 5000mA, in increments of 50mA. Press ENTER to confirm the chosen setting. The uses for this feature are as follows:
a. This can be used as a safety feature. If, for some reason, a battery never reaches peak, this feature tells the charger to terminate fast charge after this pre-set level of input capacity has been delivered to the battery.
b. Many NiMH cell manufacturers recommend to leave a partial charge on a battery before putting it in storage for an indefinite period of time. Adjusting this capacity level setting before charging to a pre-determined percentage of the battery’s maximum rated capacity, will automatically supply partial charge to the battery.
c. If this capacity setting is too low, the battery will not likely receive full charge. Conversely, if capacity setting is too high, it may not provide ample overcharge protection if the charger fails to recognize peak charge on the battery.
d. To bypass the capacity feature, set the capacity level to maximum, or 5000mAh. Caution must be used in doing so as the charger must not be left unattended.
e. If the battery’s rated capacity is not known or listed, consult your supplier to get the exact specifications for the battery.

4. The “CURRENT” setting determines how much current is to be delivered to the battery during peak charge. Press (+) or (-) to choose the desired charge current in amps. This selection ranges from 0.1 amp (100mA) to 5.0amps, in increments of 0.1A. Once the desired charge current is selected, press ENTER to confirm the setting.
If the 5.0A selection is exceeded, the display will advance to the “Auto” setting. Here, the charger will detect the internal resistance of the battery being charged every 1 minute and automatically adjust the charge current to properly match the battery. In addition, while in the “Auto” setting, if the battery type (selected earlier in the instructions) was set as “NiCd” the charger will automatically set the peak detection sensitivity level (see below) to 3mV. If the load battery selection is “NiMH,” the charger will automatically set the peak sensitivity level to 5mV.

Note: When 110V AC input power is used, the maximum charge current will be approximately 3.0A even if the display setting for charge current is set above 3.0A. Full charge current can be delivered with proper DC input power.

WARNING! Do not set the charge current too high for the battery to be charged! This could result in overheating of the battery, and lead to bodily harm or permanent damage to the battery and/or the charger.

5. The “Δ PEAK” or “delta peak” setting determines the charger’s sensitivity level for detecting peak charge of a battery. This is sometimes referred to as a “threshold” setting. Press (+) or (-) to select the desired peak detection sensitivity. This selection ranges from a high sensitivity setting of 3mV/C (milli-volts per cell in the pack) to a low sensitivity setting of 20mV/C. A higher sensitivity setting (lower numerical value) instructs the charger to be more precise when detecting peak. This is often most accurate when charge current is set at a low rate. If charge current is set at a high rate, the charger could possibly terminate fast charge too soon as voltage fluctuations on the battery and power source are more dramatic at high charge rates. A low sensitivity setting (higher numerical value) instructs the
charger to be less precise when detecting peak. This means the battery will likely be less full when fast charge is terminated, but the charger is less likely to prematurely terminate fast charge as a result of possible voltage fluctuations. A low sensitivity setting is usually unnecessary when a low charge current is selected, but could be more beneficial when a high charge current is selected. Once the desired peak sensitivity selected, press ENTER to confirm the setting. The display will return to the default display at this time.

We recommend using a setting at 6mv/c for NiCd and 8mv/c
for NiMH.

 

[Sharkonwheels]

<Getting messy and large>

For those that want to know more, here's a direct link toy Battery University's site, specifically to Nickel-based cells (NiCD and NiMH):

http://www.batteryuniversity.com/partone-11.htm

It probably goes into more detail than most want to know, but it explains the difference in charging rates, peak detection, etc...
If you check it out, look near the end at "Simple Guidelines," where they mention an NiMH cell cannot be simply used to replace a NiCD cell,
and right above it where it says it's nearly impossible to slow-charge an NiMH cell because of delta peaking on an NiMH cell.


T

 

[MikeS]

Well, call it nit-picking but I thought the difference between milliamps and milliamphours was important enough to mention; to talk about charging a battery at x mah is misleading at best (obviously if you charge your original dead 2AH NiMh at 1/2 its rated capacity i.e. 1A for an hour as 1000mah suggests, it'll be a good trick to then get 2A out of it for an hour).

And although many chargers do indeed charge at a certain rate for a certain time, this is not always optimum since different cells have different characteristics and, more importantly, there is probably still an unknown charge in the cell when you start and therefore the time to full charge is also unknown. Much better to monitor voltage, current and temperature instead of using a fixed time period, *especially* with NiMhs and LIons.

The memory effect is still being debated, but I'm inclined to agree with the opinion that it is not so much a function of the cell itself, but rather a result of many incorrect charging cycles, which applies to both NiCds and NiMhs.

m

 

[nige the hippy]

What Mr shark is saying is all true,
but pretty well every passive charge circuit I've found on an old computer has been a "trickle" charger, and neither NiCds nor NiMH mind being charged at C/10 which is a fairly usual rate for trickle charging.

I'm not sure about the HX-20, but (just looking around me) both the Gridpad, and the Thorn-EMI Liberator, just sit there pushing a trickle charge into the batteries all the while the machine's on external power.

Prudence and a meter are generally called for!

 

[Sharkonwheels]

The problem will be even worse for slow-charging, as battery university points out. At lower C rates (charging rates), it in nigh impossible to detect when NiMH cells are near capacity, and the charging may never stop, which would be a problem.

Bottom line:
I have stacks of NiMH cells, and NiCD cells.
When it came time to replace my NEC PC-8201a cells, I used NiCD.
I DID build an NiMH pack, but I did NOT want to deal with the hassles of having to measure the charging rate, and watching a clock to make sure the cell's capacity was not exceeded.

Word to the wise:
If you DO replace internal computer NiCD cells with NiMH, be wary of what the charging rate of the interna charger is, and calculate what time it would take to max-charge, and you will also need to be wary, as Mike pointed out, of the remaining charge in the packs.

I guess, when all is said and done, it ain't worth the hassle - there are NiCD cells that have higher capacities than the 20-year old cells that came with these, so that would be a better bet. Hopefully the charging hardware is smart enought to know - it should be, or else it would not be able to deal with partially-charged packs!

If you guys need links, let me know, as I have a few places I've bought HiCap NiCD cells from , and have had no issues.

T

 

[MikeS]

What Mr shark is saying is all true,
but pretty well every passive charge circuit I've found on an old computer has been a "trickle" charger, and neither NiCds nor NiMH mind being charged at C/10 which is a fairly usual rate for trickle charging.

I'm not sure about the HX-20, but (just looking around me) both the Gridpad, and the Thorn-EMI Liberator, just sit there pushing a trickle charge into the batteries all the while the machine's on external power.

Prudence and a meter are generally called for!

--------
I'd agree with that, although I prefer two meters (V & ma) ;-)
Nevertheless, I think constant trickle charging does reduce capacity somewhat, and an intelligent charger and periodic "refreshing" are best (The newer laptop batteries have most of the intelligence built in).

For a different perspective of the memory effect:
http://www.repairfaq.org/ELE/F_NiCd_Memory.html

m

 

[MikeS]

Another trick that will restore many "dead" NiCds that won't take a charge is to "zap" them with a brief high-current pulse to blow away the crystals that are shorting the electrodes. I use a 140,000 uf filter cap charged to 15V, which does the job in all except the worst cases.

Note that the cell in question will probably develop the same problem again, especially if you allow it to discharge completely, but if you keep it charged and recharge immediately after use it may well remain useful for quite a while.

mike

 

[Terry Yager]

Doesn't much matter with the HX-20, it doesn't shut down the charging circuitry, it ain't that 'smart'. Epson leaves it up to the user to remember to unplug it after 8 hours, no more, no less. It ain't designed to run from the adapter, as this will overcharge the cells. I'd imagine that NiMh cells would work, but have to be charged for a longer time. I still say, 'what's the point', the NiCads for it are cheap, easy to install, and have a decent operating time, so why take chances?

--T

 

[Sharkonwheels]

I dunno about that link, as I have been through propably near 800-1000 NiCD and NiMH cells since early 90's, and I have experienced the memory effect.
Most times it can be removed, to a certain extent, with zapping as MikeS entioned, and I have a zapper JUST for that purpose. Another trick is before the absolute first use, zap the jiminy crickets out of them, and then cycle 'em a few times. Full charge at max C, and then discharge, and repeating.

I currently have at least 60-80 cells in various states around the house/garage, and if not properly cared/maintained they WILL go south and WILl grab a "memory," although in most instances, if not harshly treated/neglected, they crystallization can be removed.

You would be surprised at the level of sophistication of some of the charging/discharging equipment used to Radio Control cars - it is utterly amazing. I have one device, that does a 5-second 30A discharge burst, waits about 10 seconds, then another, etc... until the pack reaches a level you set, and then you can download the discharge profile. It's sweet! I mean, my chargers can discharge up to like 5A, but that's nowhere near real-world use on the super-high current draw motors. 20-40A is not unheard of using NiCD/NiMH, and using some of the extreme speed controllers and brushless motors, over 2-300A is not unheard of with LiPO/LiIon and the A123 cells!


T

 

[Sharkonwheels]

Doesn't much matter with the HX-20, it doesn't shut down the charging circuitry, it ain't that 'smart'. Epson leaves it up to the user to remember to unplug it after 8 hours, no more, no less.
--T

T - what about the PX-8? Is it any smarter? I haven't looked....

T

 

[MikeS]

Some of my stuff that uses batteries doesn't have any charging circuitry anyway (M100, MP3 players, Boom boxes etc.), but even with the stuff that does (Scanners, ham radio portables etc.) I just keep several spare sets charged with an intelligent charger and just swap them as needed, mostly NiMh these days. The only annoying problem is that equipment meant for primary cells generally shuts down too early because of the lower voltage of the rechargeables.

m