Low self-discharge NiMH batteries for the Yaesu FT-897

Internal batteries in the FT-897

I liked operating portable with my FT-897 with its internal FNB-78 battery packs, but those NiMH packs did not work well in my infrequent use. When they finally died, I installed lithium polymer batteries in their place. They worked wonderfully, but I may have had a bad batch of cells, because they puffed up after a very short period of very light use. I was left wishing for a perfect battery chemistry that would have low self-discharge, high capacity, low internal resistance and better behavior than those RC model LiPo cells.

Well, there is a perfect battery chemistry: the Low Self-Discharge NiMH pioneered by Sanyo with its Eneloop brand, now owned by Panasonic. That's what I use exclusively in all my AA battery applications such as cameras, flashes, PMRs, scanner radios, antenna analyzers, etc. Those cells really do not discharge even during many months of storage (up to a year or so is the longest I have experience of, and they still held plenty of charge), their internal resistance does not go way up with disuse, and they retain all of their rated capacity even after years of use and disuse. Ordinary NiMH batteries are a disaster in comparison, and I haven't bought a single one of those since I discovered Eneloops! If only they made cells in the 4/3 A size (the common size used in battery packs, including Yaesu's FNB-78), I would have rebuilt my batteries with those and never looked back! But although nowadays there's plenty of manufacturers of similar low self-discharge NiMH batteries (GP ReCyko, Ansmann Max-E, Duracell Precharged, Powerex Imedion, and lots and lots of various unknown brand names), no-one seems to make a 4/3 A cell. Hello, why not??? The biggest market seems to be for AA and AAA cells only, and some manufacturers even make D cells of ridiculously low capacity, which are actually just single or dual AA cells in a D cell sized package. While AA size Eneloops probably could stand up to the transmitter's power requirements, their 2 000+ mAh capacity is just too small for my taste.

External batteries for the FT-897

So after a few years' hiatus from portable ops, I finally decided to build an external battery pack using low self-discharge NiMH D cells made by Ansmann under the brand name "Max-E". These cells provide 8 500 mAh of capacity (which I verified individually before soldering them together, and they do come very, very close) and have no problem with the 3 or 4 amps that the FT-897 maxes at when transmitting SSB at 20 W output (the consumer D cell is rated for 8.5 A continuous discharge according to the datasheet, the industrial version much higher still). They are sold in blister packs of two cells each, so six packs are needed for the 11-cell (13.2 V) configuration used in the FNB-78, with one cell left over. Since I'll be making another 11-cell pack also, that odd cell won't go to waste.

These 34 mm diameter cells do not, however, fit inside the battery compartment of the FT-897, which is only some 26 mm high. Other DIYers have fitted their FT-897s with a new battery compartment cover that extends outwards from the bottom, making room for the fatter cells. I decided instead to forgo the convenience of internal batteries, and make an external pack. After all, that could be used to power other devices as well, so it also has its advantages. But I do sorely miss the internal batteries...  :(

After verifying their capacity, I glued the cells together with hot glue, just to keep them in formation while I solder them. One of the biggest limitations of hot glue is its tendency to melt in high heat (well, duh), but I figured the cells should not get really hot in my use (i.e. nowhere near the 120°C that is often quoted as the melting temperature of hot glue, though it does soften already much below that), and I'd be wrapping the whole thing in heat-shrink tubing anyway (being careful not to melt the glue in the process, and especially not to destroy the thermal fuse I installed).

Hot glue is not 100% permanent (I consider that an advantage here), i.e. they can be torn or cut apart after gluing, but regardless, I did draw a diagram of the orientation of each cell, and how I intend to connect the cells, the terminal wires, and the thermal fuse together. I checked and re-checked this diagram before beginning the assembly process, and checked and re-checked again before gluing each cell. And I kept checking it while soldering (see below), just to be sure. And hey, I got it right the first time, so it all paid off!  :)

Next I soldered the cells in series. For the interconnects, I used 8 mm wide nickel strip that I ordered from eBay. With a touch of good quality flux, I first pre-tinned both the battery terminals and the strip, and then just melted them together. I used my 80 Watt Weller WD-1000T soldering station with a very wide 4.6 mm LT-D tip, set at a rather high 380°C temperature, in order to melt the solder as fast as possible. A lower temperature, a narrow tip, or an underpowered soldering iron would require more heating time, actually allowing more heat to flow into the cell. Heat damages the cells very easily, and various sources on the Net tell you not to solder to NiMH cells, but to use spot welding instead. (In fact, that's the purpose this nickel strip is intended for, not soldering.) A DIY spot welder should not be too difficult to make, but the videos I've seen on Youtube don't really make it look all that gentle either. Anyway, my cells survived soldering just fine.

For safety, I inserted a 72°C thermal fuse (the thing with the yellow wires in the photo, enclosed in black heat shrink tube) in the space between the middle three cells of the pack. It is only meant to break the circuit in case the charger goes berserk and starts to overcharge the battery pack for hours on end. I chose a 72°C fuse because that was the lowest temperature available—it's still high enough to allow charging the batteries with reasonable currents. I crimped the wires to the fuse, enclosed it in heat-shrink tubing (which I shrunk only at the ends to avoid damaging the fuse) and soldered the wires in series with the cells.

The 11-cell packs (I first made one for testing, then another identical one) were enclosed in PVC heat-shrink tubing, which can be shrunk with an ordinary hairdryer. (Many commercial battery packs as well as lithium polymer batteries are enclosed in the same type of material.) Since it does not require extreme heat to shrink, it is safe for the cells, as well as the hot glue keeping them together, and the thermal fuse amidst the cells. But it did turn out ugly as hell.

The battery packs' terminal wires were soldered directly to the cells, and 2-pin Molex connectors were attached for good measure. I could just as well have wired them directly to the rest of the circuit (see below), but a connector seemed appropriate in case I need to remove or replace a pack. I could have used Powerpole connectors here also, but somehow I never considered them for this kind of internal connections. Damn. I did, of course, use Powerpoles for the external connectors of the battery box, but there's really no reason not to use them internally as well.

The two 11-cell battery packs are wired to three Powerpole connectors according to this wiring diagram. Each battery has a dedicated connector with its own on/off switch, and there is also a shared connector that can connect to either battery via a third on/off/on 3-position switch. This allows:
  • two packs to be used completely independently,
  • one pack to be in use while the other is charging,
  • two loads to be connected to a single pack without needing a splitter cable,  and
  • immediate switching of the radio from one battery pack to another with a single switch.
That last feature is the most important in practice, and does what the "A/B" switch on the radio does for internal batteries. A single returning (on)/off/(on) switch wired to a cheap 2-wire volt meter ordered from eBay allows checking the voltage of either pack at any time (although the FT-897 also shows the battery voltage on its display at all times).

All switches are two-pole, so they isolate both the positive and negative side of the pack. In some chargers, e.g. my Bantam e-Station BC6-DC, the negative output lead is not at ground potential, which would cause a fuse to blow if the pack's negative side was grounded via the radio. The two-pole switches mitigate that problem entirely. Also, they enable connecting the packs in series, in case a 24 V portable supply is ever needed. Finally, each battery pack has its own 5 A automotive-type blade fuse. The fuses are held in a four fuse holder, which leaves two unused slots to keep spare fuses in, which is nice.

The packs are housed in a plastic food storage box with a lid that can be locked in place. I fitted two horizontal dividing walls inside. The upper one has all the connectors and switches installed on it (as detailted in the schematic above), while the lower one provides a smooth surface against the battery packs, which are housed in the lower part of the box. The bodies of the connectors and switches, as well as all the wiring, are in the space between the walls, which are separated by spacers made of 25 mm square aluminum tubing. The original lid of the box still fits on top for transport, in order to keep dirt out of the connectors and to prevent the switches from turning on inadvertently. The fuses are accessible with the lid open, in case they need replacement.

The whole thing weighs about 4 kg (about the same as a 12 V 12 Ah lead-acid). The box and the dividing walls are both translucent or transparent plastic for no good reason, which makes all the uglyness inside visible to the whole world. Whatever, as long as it works... Simply remove the lid, attach the radio's power cable, flip a switch and start operating.

The FT-897 is designed to accommodate two internal battery packs, which connect to two male 3-pole JST VH headers with 3.96 mm pitch inside the radio (you can find compatible connectors on eBay—search for "JST VH 3p 3.96"), and either one can be taken into use with the "A/B" switch on the radio. I wired both headers to separate Powerpole connectors which I mounted on the back of the battery compartment cover with a makeshift arrangement of aluminum profile, cable ties and hot glue. Now I can connect my external battery pack to one, and another power source (another battery, a car 12 V system, or an AC/DC adapter) to the other, and select either one with the radio's "A/B" switch.

The reason why I don't just wire the pack to the radio's external 13.8 VDC input is that when powered via the internal connectors, the radio automatically limits its TX power to just 20 W, without needing to change the power setting through the menus. And I definitely want the lower power when operating on batteries! (Incidentally, the FT-857, which is like a smaller FT-897 without provision for internal batteries, uses the extra pins on its 13.8 VDC input connector to control its output power: if pin 3 of the attached power supply cable is wired to GND, TX power is automatically limited to 20 W. The FT-897 obviously does not have this feature on its external power supply connector, since there is physically no pin installed at the pin 3 position of the connector. Bummer.)

Charging the batteries

I use a Bantam e-Station BC6-DC to charge the batteries. It is a versatile multifunction charger that can charge, discharge, balance and condition NiCd, NiMH, Li-ion, LiPo, LiFe and SLA batteries of numerous cell counts at adjustable currents up to 5 A. (The e-Station BC6 is the same thing with an internal AC adapter, and the BC6-10 has a maximum charge current of 10 A; otherwise I think they are the same.) When charging NiMH batteries, it is advisable to change the "NiMH Sensitivity D.Peak" setting in the "USER SET PROGRAM" menu from "Default" (8 mV/cell) to its minimum value of 5 mV/cell. Also, set some reasonable values for "Capacity Cut-Off" (I use 10 000 mAh for this 8 500 mAh pack, and have never had it cut off before completion) and maybe "Safety timer" as well (at 2 A current I use 400 min, which is plenty).

For these 8 500 mAh cells even the maximum current of the BC6-DC is less than 1C. The Ansmann Max-E datasheet doesn't specify a maximum charge current, but quotes charging times for 2.55 A (0.3C) and 4.25 A (0.5C) currents. Since I'm usually not in a big hurry, I generally charge at just 2 A. That's about the same 0.25C that I like to use for my 2 000 mAh AA Eneloops when charging on my TechnoLine/LaCrosse BC-900 charger (they get hardly warm at 500 mA current, whereas at 1000 mA they do heat up substantially towards the end), and those have served me for years on end. With any luck, and with conservative charging, maybe these D-cell battery packs will as well. (But with a single NiMH cell, the BC6-DC doesn't seem to reliably detect the endpoint at 0.25C, and was inclined to overcharge. So while I was testing the individual cells, I used 0.5C current i.e. 4 A, and the cells did become quite hot towards the end. Hot enough to soften hot glue, in fact. But with the whole pack, the BC6-DC terminates just fine at 2 A, and the pack hardly gets warm at all.)

The new batteries in use

The original Yaesu battery packs were rated at 4 500 mAh each, and lasted a weekend's intermittent operating with 10 W SSB just fine ("intermittent" means whatever time is left over from fishing, eating, sauna, sitting infront of the fireplace, etc... which is, on the whole, not a lot), but their voltage was always drooping during transmission even at low power. The lithium batteries I replaced them with were an absolute dream while they still lasted: 6 600 mAh capacity each, and no voltage droop whatsoever even at 20 W SSB. These external NiMH battery packs are even higher capacity, 8 500 mAh each (almost twice the original Yaesu ones) and work equally well as the lithiums, but of course aren't internal to the radio, which is a pity. Let's hope they last longer than the lithiums did.

Since the FT-897 is designed to operate on 11-cell NiMH packs, with a nominal voltage of 13.2 V (in practice the voltage ranges from a maximum of 16.5 V to a minimum of around 11 V), its LED indications of battery condition work as intended: the LED shines steadily down to 12.4 V, then begins to blink, and turns off completely at 10.8 V. I stop using the pack when it reads 12 V during RX, or if the LED starts to turn off completely during TX. After that there's very little capacity left unused, and no point in pushing the batteries any further.

Now that the battery compartment is empty, I wonder if I could fit my LDG Z-100Plus Autotuner there in some sensible way; that might be quite convenient also... Or I could wire up a single battery pack of 11 C cells in there (they do fit if they're not surrounded by extra packaging). But a single 4 500 mAh battery pack alone (that's the rating of Ansmann's Max-E C cells) would be disappointingly small for portable ops, and it would hardly add much to the 17 000 mAh total of the external packs.

A final warning

It is common knowledge that lithium batteries can catch fire or even explode if misused. However, NiMH batteries also pack a lot of energy in a small volume, and have a very low internal resistance. Even if it doesn't contain as much flammable material as a LiPo cell, a shorted NiMH will get very hot very fast, and can cause a fire hazard regardless. Treat them with respect. Don't puncture them, don't drop them, and make sure no short circuits are caused by exposed connectors, or loose or frayed wires. Don't leave out the fuses, no matter how competent you think you are.

In case you're not already proficient at soldering, NiMH cells are not a good place to practice! You may damage the cells, or even have them explode in your face. Professional battery pack builders really do use spot welding instead of soldering! But even if you do choose the spot welding option, do your homework first! If you don't know what you're doing, you may still damage the cells, or have them explode in your face!

Of course, NiMH batteries should only be charged with a dedicated NiMH charger. Don't try any other charger—not even your grandfather's old RC hobby NiCd charger (although NiCd and NiMH have similar endpoint detection, a NiCd charger is generally not sensitive enough to detect the smaller negative delta V cutoff point of NiMH batteries). And never charge any batteries unattended.

Read more about NiMH batteries at Battery University. And if you choose to try this yourself, you do so at your own risk. And please note, that anyone who talks of a device consuming "xx mAh/hr" or "xx amp-hours per hour", or needs to google "convert hours and amps to mAh", probably should not be involved in anything DIY with rechargeable batteries. And anyone who recommends charging a given battery "at a rate of xx mAh", or uses units of "A/h" or "mA/h" in any context, should definitely not be let loose around batteries!!! (Ask me why.)

You have been warned.

Antti J. Niskanen <uuki@iki.fi>