Faceoff: GP3300 vs. 4th Generation Kokam

It has been over a year since I did the last batch of tests on LiPo cells. At that time, a significant shortcoming noted was the inability to supply high discharge currents. The work-around was to parallel enough cells to provide the desired current. This resulted in very expensive packs that lacked the reputed light weight of LiPo, but with obscenely high capacity. My main interest is in sport flying of "40-size" and larger planes that need 30+ amps to fly well. I finally replaced the 2 10xCP1700 packs for my Sukhoi with the 2 3S4P Thunderpower packs I bought for last year's tests. The CP1700's gave a 7-minute aerobatic flight. The TP were the same weight as the old packs, but now I can fly four 8-minute flights on one charge. This is nice, but I'm not sure it's $500 nice.

That was then, this is now. FMA Direct has introduced their new "4th generation" Kokam packs. This includes cells of 2100 MAH and 3200 MAH rating with a discharge rating of 20C continuous! This means that you can discharge a pack safely in 3 minutes! Comparing these cells to previous generations, the new cells are larger and heavier and more costly. Maybe so, but these new cells get the job done. The real comparison is with the ubiquitous NiCad and NiMH cells. They've been doing the job for years.

While testing the new Kokams, I decided to test some of the newer NiMH cells. Early NiMH were disappointing, with much less than advertised capacity at our discharge levels, and higher internal resistance than comparable NiCads. The GP3300 cells are representative of the improvements that have been made, and have proven to be quite popular. I procured two 6-cell packs (my friend Bill is now into electric racing boats and 12 cells is the most popular configuration). I was impressed by these cells. They had the lowest resistance and highest capacity of any Nickel cells I had tested. What struck me, though, was the similarity of form factor to the KOK3200-20C-2S pack that I was testing. My Thunderpower packs, even though the same weight as the 10xCP1700 packs, are much bulkier. But I digress. On with the Faceoff! GENERAL Both NiMH and LiPo packs have long charge times compared to the 20-minute charge for NiCads. Some charge NiMH at higher than 1C rate, but conventional wisdom is that this reduces useful life. LiPo have the advantage of not losing significant charge when not used immediately after charging. Ni* cells not only lose charge, but lose capacity if not used regularly. The capacity can be restored through charge/discharge cycles, but this is a real pain. These LiPo cells promise over 500 cycles of useful life. Experience shows that Ni* cells don't hold up that well at our levels of abuse. LiPo have special handling considerations that, if not heeded, will result in premature failure. FORM FACTOR Remember the basic rule-of-thumb that one LiPo cell is equivalent to three Ni* cells? The KOK3200-20C-2S (2S1P) is somewhat narrower, shorter, and thinner than the equivalent 6-cell GP3300 pack, as you can see below.

This is a good start. We can see that the Kokam pack is basically a drop-in replacement. The 3S pack basically stacks another cell on top, making it slightly higher than the GP pack, but definitely shorter than a 9-cell pack. WEIGHT No contest here. The Kokam pack weighs HALF as much as the GP3300. As you will see below, using Kokam in place of GP reduces my eNobler from 100 to 79 ounces! LAB TESTS First, let's compare voltage. As we've done in the past, the graphs are normalized with a ratio of 3:1 to compare NiMH and LiPo.

The Kokam cells are slightly lower voltage than the GP cells, but keep in mind that the GP have the highest [open-circuit] voltage I've ever tested. What matters more is the expected output voltage under load. We measure open-circuit voltage and internal resistance in order to properly model the cells in ElectriCalc. The following graph (again normalized 3:1) shows the real differentiator of the two packs.

The normalized resistance of the Kokam cell is HALF of the excellent resistance of the GP cells. This means that for 20A+ loads, the Kokam output voltage will be higher and the internal losses [heat!] will be half of the GP. This is major important. As an aside, note that the Kokam resistance decreases with discharge. This seems common with LiPo, and apparently is a function of the cell warming up. The upsweep on the GP curve is typical of Ni* and contributes to the "brick-wall" effect at the end of discharge.

PERFORMANCE ESTIMATE I took my scratch-built eNobler and did a quick ElectriCalc comparison between my CP1700's and the Faceoff cells. The eNobler has a 550 sq.in. wing area, a MaxNEO-13Y motor and 14x7 APC-E prop. I did not change the prop, but reduced the cell count from 20 to 18 and rule-of-thumb equivalent 6 LiPo. This acknowledged the higher internal resistance of the CP-1700's. With their greater weight, the GP3300 did not yield better performance, save for longer flight times. Arguably I could have increased prop size to increase performance, but this comparison highlights the superior internal resistance of the Kokam cells. Along with the weight came undesirable high wing loading. The Kokams, on the other hand, make everything better. The results below show the impressive performance of the Kokam cells.

PRICE Finally the GP3300's win one. The KOK3200-20C-2S sells for $98.50. I paid $39.95 for a pack of 6 matched GP3300 cells. I have seen unmatched packs for $26.95. Is the performance worth the price? Arguably, for my eNobler the extra cost amortizes to about 25 cents a flight if I don't crash them or screw up the charging. Further rationalization argues that the lower wing-loading will keep me out of trouble and extend the life of my plane. Hopefully, as this technology becomes mainstream, prices will drop. CONCLUSION If money were not a consideration, I'd have the Kokams in all my planes and enjoy a substantial performance gain. Alas, I still have to make cost/benefit decisions. Such is life.

ECALC DATA: CELLS MAH mOHM DIA. HGT. WGT. MAH/OZ. COMMENTS (typ) (in) (in) (oz) KOK640-15C 580 53.1 1.40 2.50 0.65 892 3.75 0.19T LiPo 9.6A cont. (FMA-Kokam) KOK910-15C 850 22.5 1.48 2.57 0.85 1000 3.75 0.21T LiPo 14A cont. (FMA-Kokam) KOK1250-15C 1160 22.5 1.68 3.18 1.35 859 3.75 0.23T LiPo 18A cont. (FMA-Kokam) KOK2100-20C 1885 12.5 2.55 4.35 2.75 685 3.75 0.32T LiPo 42A cont. (FMA-Kokam) KOK3200-20C 2990 7.5 1.72 5.22 3.20 934 3.75 0.32T LiPo 64A cont.! (FMA-Kokam) GP3300* 3210 5.0 0.91 1.69 2.20 1459 1.25 NiMH

If you have V2.2, you can load new database entries automatically into ecalc by clicking File then Merge Files. You can download a free upgrade if you have V2.0 or V2.1, or pay 10 bucks if you still have V1.x.