docholliday
Well-known member
TL;DR/Lesson of the story: Buying a good charger makes the difference; Watson (and Hahnel to a less extreme) chargers suck.
Over the last two weeks, I've been exploring battery issues with the Canon style BP-9xx cells used by Phase One in their digital backs. The primary issue was random weird effects such as new batteries exhibiting unusually short battery life and very old batteries showing good life. Some of the old batteries are 8-10 year old 2900mah batteries and they were outlasting both new Phase One "OEM" 3400mah batteries as well as the Capture Integration Certified "non-OEM" 3500mah batteries.
An email conversation with P1 support yielded not much help other than hearing "we don't recommend those CI non-OEM batteries" which is a absolutely ridiculous thing to say as these are non-smart batteries (they don't have monitor chips). But, just to be sure, I purchased a brand new, official P1 battery and it suffered from the same problems. So, I decided to do my own testing and diags. Having both a Hahnel Twin-V and Cube charger, as well as a Watson, I was looking at battery charge levels as a possible cause. I also have a LiON cell analyzer, electronic load, and other test equipment, so it made sense to not trust other's diagnosis of the issue.
I noticed that the battery levels shown on the back and on each of the different chargers were rarely the same. Off by a bit is understandable, as again, these aren't smart batteries that have a monitor IC on the battery to store the data. Each device had it's own way of determining the charge level. One thing I noticed consistently was that when the back would show a dead battery (at the point of shutting off), the Hahnels would show a dead battery, but the Watson would show 20-30%. Further investigation showed even crazier discrepancies. I tried each the new P1, one of the CI-Certifed, and one of the old 2900mah batteries in the back after a full charge (100% on display + another 30-45 mins) and then leaving the back on until it showed 70%. Putting the batteries into each charger showed: 70 or 80% on the Hahnels and 100% on the Watson.
The LiON analyzer showed that all the batteries were fairly similar in performance. Measurements were fairly similar between the new P1, CI, and old battery. Each battery fully charged showed around 8.2-8.4v, which is correct. The 7.2/7.4v marking is a nominal rating, and these batteries can run as low as 6.4v when drained. Most of these batteries are constructed of 18650 or 18500 cells in series, but without a tap on the center link, a cell analyzer can't do much to check the cell balance or per-cell performance.
I ended up resolving the issue by purchasing a Dolgin Engineering TC400 charger for the Canon BP-900 batteries. Their charger can rapid charge 4 batteries simultaneously but more importantly, uses it's own charging algorithm and also has both a charge-discharge-charge "TDM" capacity test mode and a display of the total charge pushed into the battery upon completion.
I started by cycling each of the batteries on the TDM test mode. Each battery reached its expected nominal level or slightly exceeded it except for the old battery which, as expected, had wear but still took 2600mah of charge. Those are some robust cells! This confirmed that the batteries were fine, even the non-OEM ones. The new P1 battery was cycled a half dozen times to form the cells before I started testing.
As a side note, the biggest problem with the Twin-V wasn't with the charger electronics, but the design of the removable plate. The contacts are simply solder coated and over time, oxidation makes the contact between the charger and battery plate fail. Cleaning and coating with DeOxit D100L + Shield fixes the issue of the charger not recognizing the battery. Or in my case, I simply took the charger apart, ripped the spring contacts out of the charger and soldered wires (with non-RoHS compliant leaded solder) directly from the board to the plate. But, it was still causing inconsistent charges, albeit not as poorly as the garbage Watson charger. I have a feeling the Cube will suffer eventually from the same problem as it uses a removable plate system too.
A "fully" charged battery on the Hahnel ("fully" is in this case, waiting for 100% to appear on the display and giving another 30-45 mins) would show 80-90% (and occasionally 100%) on the Dolgin. Here's the new P1 OEM battery after fully charging:
Transferring that battery over to the Dolgin showed the following:
The back would show 100% for a few minutes which rapidly dropped to 90%. Even worse, a fully charged battery from the Watson showed 70-80% on the TC400. With these batteries, letting the TC400 top off the battery until it read "Completed" resulted in a battery that held its charge more accurately. Using the Dolgin to charge depleted batteries exclusively has resolved the issue. All the batteries now perform in a linear, predictable fashion. It's also a bonus to see how much the battery took during charging and to be able to cycle the batteries to determine wear. The Hahnels will probably get put in a drawer as backups. And that Watson has a nice date with a burn pit or 12 gauge...
Over the last two weeks, I've been exploring battery issues with the Canon style BP-9xx cells used by Phase One in their digital backs. The primary issue was random weird effects such as new batteries exhibiting unusually short battery life and very old batteries showing good life. Some of the old batteries are 8-10 year old 2900mah batteries and they were outlasting both new Phase One "OEM" 3400mah batteries as well as the Capture Integration Certified "non-OEM" 3500mah batteries.
An email conversation with P1 support yielded not much help other than hearing "we don't recommend those CI non-OEM batteries" which is a absolutely ridiculous thing to say as these are non-smart batteries (they don't have monitor chips). But, just to be sure, I purchased a brand new, official P1 battery and it suffered from the same problems. So, I decided to do my own testing and diags. Having both a Hahnel Twin-V and Cube charger, as well as a Watson, I was looking at battery charge levels as a possible cause. I also have a LiON cell analyzer, electronic load, and other test equipment, so it made sense to not trust other's diagnosis of the issue.
I noticed that the battery levels shown on the back and on each of the different chargers were rarely the same. Off by a bit is understandable, as again, these aren't smart batteries that have a monitor IC on the battery to store the data. Each device had it's own way of determining the charge level. One thing I noticed consistently was that when the back would show a dead battery (at the point of shutting off), the Hahnels would show a dead battery, but the Watson would show 20-30%. Further investigation showed even crazier discrepancies. I tried each the new P1, one of the CI-Certifed, and one of the old 2900mah batteries in the back after a full charge (100% on display + another 30-45 mins) and then leaving the back on until it showed 70%. Putting the batteries into each charger showed: 70 or 80% on the Hahnels and 100% on the Watson.
The LiON analyzer showed that all the batteries were fairly similar in performance. Measurements were fairly similar between the new P1, CI, and old battery. Each battery fully charged showed around 8.2-8.4v, which is correct. The 7.2/7.4v marking is a nominal rating, and these batteries can run as low as 6.4v when drained. Most of these batteries are constructed of 18650 or 18500 cells in series, but without a tap on the center link, a cell analyzer can't do much to check the cell balance or per-cell performance.
I ended up resolving the issue by purchasing a Dolgin Engineering TC400 charger for the Canon BP-900 batteries. Their charger can rapid charge 4 batteries simultaneously but more importantly, uses it's own charging algorithm and also has both a charge-discharge-charge "TDM" capacity test mode and a display of the total charge pushed into the battery upon completion.
I started by cycling each of the batteries on the TDM test mode. Each battery reached its expected nominal level or slightly exceeded it except for the old battery which, as expected, had wear but still took 2600mah of charge. Those are some robust cells! This confirmed that the batteries were fine, even the non-OEM ones. The new P1 battery was cycled a half dozen times to form the cells before I started testing.
As a side note, the biggest problem with the Twin-V wasn't with the charger electronics, but the design of the removable plate. The contacts are simply solder coated and over time, oxidation makes the contact between the charger and battery plate fail. Cleaning and coating with DeOxit D100L + Shield fixes the issue of the charger not recognizing the battery. Or in my case, I simply took the charger apart, ripped the spring contacts out of the charger and soldered wires (with non-RoHS compliant leaded solder) directly from the board to the plate. But, it was still causing inconsistent charges, albeit not as poorly as the garbage Watson charger. I have a feeling the Cube will suffer eventually from the same problem as it uses a removable plate system too.
A "fully" charged battery on the Hahnel ("fully" is in this case, waiting for 100% to appear on the display and giving another 30-45 mins) would show 80-90% (and occasionally 100%) on the Dolgin. Here's the new P1 OEM battery after fully charging:
Transferring that battery over to the Dolgin showed the following:
The back would show 100% for a few minutes which rapidly dropped to 90%. Even worse, a fully charged battery from the Watson showed 70-80% on the TC400. With these batteries, letting the TC400 top off the battery until it read "Completed" resulted in a battery that held its charge more accurately. Using the Dolgin to charge depleted batteries exclusively has resolved the issue. All the batteries now perform in a linear, predictable fashion. It's also a bonus to see how much the battery took during charging and to be able to cycle the batteries to determine wear. The Hahnels will probably get put in a drawer as backups. And that Watson has a nice date with a burn pit or 12 gauge...
Last edited: