LTO (Lithium Titanate Oxide) – The Ultimate Battery for Dash-Cam parking mode (DIY)

[EricSan]

Ha- ooops, it's the Viofo HK4 that ignores the low voltage cutoff when you tie together Red + Yellow wires on the 12v input side of the regulator (not Vantrue).

When I came home from work today, the N4S 3-ch camera was no longer recording. It seems that it shut down at about 9a Monday morning. That is ~6 hours short of 9 full days of battery power (give or take a little because I experimented with turning it off and then applying charging voltage to make it wake up) before the battery hit 11.8v and the dashcam hardwire kit shut down the camera.

At 9p tonight, I powered up the N4S with the Viofo hardwire kit so it will run on a low voltage battery. I thought the GPS would allow the camera to enter parking mode after a few mins of no apparent movement, but that doesn't seem to be happening, it's not dropping down to a lower power consumption mode. Battery voltage is dropping more quickly now, so I'm not expecting the battery to last much longer than an hour or two. I want to run the battery down far enough so that the SOC-0% Volt.(V) limit gets hit so the BMS recalibrates the depleted battery voltage.

EDIT: That didn’t take long at all! The camera only lasted one more hour before the BMS turned off the output. The camera is now power cycling as the battery voltage climbs back up to the Under Voltage Protection Recovery level after being shut down by the UVP setting.

 

[EricSan]

Quadruple the storage capacity for the new battery, with a similar efficiency rating (maybe not, these are used batteries) and a slightly higher draw of 4.5w (camera + hardwire kit) + 0.7w for the BMS should come closer to 0.95*1104/5.2w = 202 hours or 8.4 full days. So, yeah, anywhere between 7 and 9 full days is my ballpark estimate for single camera in parking mode.

Don't you just love it when theory matches practice?!?!? 😉

My battery died (dashcam hardwire kit shut things down at 11.6v) last night after running for 210.6 hours (nearly 9 full days). My predicted runtime was 202 hours. Running my calculations backwards means either there is more battery capacity than 1104Wh (more likely), or that I obtained an overall efficiency of 99.2% (less likely). Either way, SUPER AWESOME!!!

Here is the battery discharge curve that I created from a series of BMS app screen captures. My timing intervals were clearly not uniform, but the curve is a near perfect match to the theoretical discharge curve for LTO cells. There is one anomaly in the data at about the 88 hour mark (this bump is also present in the manufacturer's data shown below). I think this is when I was playing around with seeing if the BMS/camera would wake up and turn on after being turned off with the BMS power button. I think I inadvertently left my external PSU connected for about 5-10mins after one of the wake ups, which accounts for the little bump in the curve. I also played around with adding extra draw with my power resistors, which is likely what brought the voltage back down again. If you ignore the one or two data points that are not in line, the rest of the curve is near perfect.

I only got another 1 hour of battery time when I bypassed the hardwire kit shut down setting (11.6v) and let the BMS shut things down at the Cell UVP Under Voltage Protection level of 1.800v per cell (10.8v for the entire battery). Conclusion (for me): the hardwire kit with the low voltage protection circuit disabled is not worth the trouble, I would only gain 1 more hour for a single camera. This would translate to an extra 15-20mins when running 3 or 4 cameras.

I also noticed that the Coulomb Counter in the BMS reset the remaining battery capacity when the voltage bottomed out. It had been reading rather high as time was going by.

Looks like those "used" LTO batteries are operating darn near 100% capacity.

EDIT: Hmmm... looking at the official Toshiba SCiB documentation concerning discharge rates for this exact battery pack, it looks like that little bump in the discharge curve at about 88hrs is characteristic of the battery construction and not an artifact of my interruption of the discharge process. Check out the blue trace (0.2C) in the chart below:

 

[GPak]

Excellent result! @EricSan , Thank you!

Just in case the coulomb counter in BMS cannot be accurately calibrated:
I divided your curve into 10 equal intervals (from 0% to 100%), starting at 12V and ending at 15.5V (assuming this "linear" range corresponds to the entire battery voltage range in a resting state), and rounded the corresponding voltage values for easier memorization, or for a cheat sheet🙂.

Here is the conversion of battery voltage (V) to battery charge level (%):
12.0 V – 0%
12.5 V – 10%
12.75 V – 20%
13.0 V – 30%
13.25 V – 40%
13.5 V – 50%
14.0 V – 60%
14.25 V – 70%
14.5 V – 80%
15.0 V – 90%
15.5 V – 100%

Or a simpler option, excluding the double decimal voltage values highlighted in orange:
12.0 V – 0%
12.5 V – 10%
13.0 V – 30%
13.5 V – 50%
14.0 V – 60%
14.5 V – 80%
15.0 V – 90%
15.5 V – 100%

EDIT: in blue (based on edit of above post)

 

[EricSan]

Great chart, @GPak ! I was contemplating mapping out the same set of voltages and charge levels so I could have a more clear understanding of how things work - thanks! I would agree, the "effective" voltage range of this LTO battery 12.0 to 15.5 (I need to reset the limits on my LED battery meter accordingly). Once the battery dips below the 12.0v mark, there is only 3-4 hours max left with the draw of a single camera. Despite my uneven timing intervals for voltage measurements, that's a pretty linear draw-down curve.

I made two more discoveries last night as I was playing around with the Smart Sleep function:

1) In terms of "wake up" behavior, this BMS wakes up from being put to sleep with a manual press of the the power button and from the automatically induced Smart Sleep inactivity timer in exactly the same manner when a charging voltage is applied. It does not seem to distinguish HOW it went to sleep - it just wakes up, starts charging, and the camera cones on right away! Awesome! This is exactly what I was hoping to find.

2) I've been testing the battery with a Vantrue N4S 3-ch camera that provides a 4.5w draw in LBR parking mode. This power draw is not sufficient to keep the BMS from going to sleep when the Smart Sleep function is enabled (bummer). However, when I added a second camera to my battery terminals (a Vantrue S1 Pro Max without the rear camera), the combined power draw of these two cameras IS sufficient to keep the BMS from going to sleep! More Awesome! In an interesting observation, the combined power draw of these two cameras as reported by the BMS is 7.2w or 0.57A - exactly the same figure you noted as the minimum current draw needed to calibrate the ammeter. It seems that we've converged on the "minimum" threshold that the BMS likes to see for proper operation from two different directions. Since my battery is mostly discharged at this point (12.5v) the ampere reading is a bit higher (0.57A) with two cameras running. I'm wondering if these two cameras will STILL be enough to keep it awake when there is a higher state of charge (say, 15v) and the corresponding current draw will be lower than 0.57A... So, I'm wondering exactly "what" gets counted by the BMS in order to keep it awake: the current or the wattage draw? I'm guessing that it's current draw (easier to measure without a calculation) and that these two cameras won't hit the necessary threshold when the battery is more fully charged. Guess I'll find out soon enough as I charge the battery.

Next, I want to charge the battery in a more "controlled" manner (not all in one shot), so I can stop along the way and investigate the BMS "wake up" function at specific voltage intervals. I think I'll use the 10% increments that you posted above and see how things go. I'll have to monitor things pretty closely so I don't overshoot each 10% charge increment too much. It takes far too long to draw down the voltage level when you only siphon off a few watts at a time (each half of a volt drawdown takes nearly a full 24 hours). My big battery tester is supposed to arrive later this week or early next week, so that should speed things up a little for testing purposes.

 

[GPak]

@wilsch
Long time no updates, how is your battery working?
It's quite cold in Canada right now.

@viciouslancer
Have you finished assembling your battery?
It would be interesting to hear about your experience using 10Ah cylindrical LTO cells.
I'm thinking about buying 6 cylindrical 10Ah cells and assembling a relatively small, about 140 Wh battery.
I have all the necessary spare components, need to use them.

 

[EricSan]

Had to add a high current negative terminal that sits on the "other side" of the BMS so I can use the Lithium Battery Tester to draw down the battery and have the BMS be aware of the current draw. I picked up a pair of terminals on AliExpress for $5. All of the same ones on Ebay are going for $25/pr. These are the "small" ones, rated at 120A - a bit of overkill, but nice for the price and does the job.

https://www.aliexpress.us/item/3256807290521655.html

I just connect the other clamp from the battery tester to the positive battery terminal. At first, I was getting frustrated that the battery tester wasn't functioning properly because it was turning itself off after only a few seconds of discharge. Then I realized it was the BMS setting that I had changed that was shutting things down - oops, another case of "operator error."🤣

It took 9 hrs to fully recharge the battery at 10A charge rate 🫨. Doing the math on an 80Ah battery, a 1C charge rate is 80A. A 10A charge rate in my car works out to a paltry 0.125C, while a 0.75A discharge rate from powering three cameras works out to a 0.01C discharge rate 🤣. Thus, it looks my set of batteries will be used rather gently, making me think I'm likely to get a full 20 year service life from them...

The high current terminal comes with a plastic slip cover so the terminal isn't always exposed. When I'm done experimenting with the battery, I'll remove the screw and install the terminal cover.

Now that I have the ability to draw down the battery level faster than a 5w camera can, I want to see if the battery is able to wake up at different State of Charge levels. On the plus side, the BMS did put itself to sleep after fully charging the battery and then just sitting there for an hour, which is pretty cool. As I observed before, the BMS starts ramping down charge current when it gets within 1.0v of a full charge. It charged at 10A for about 7 hours before it started ramping down the current. From there, it was another 2 hours until the battery was full.

With a 10A charge rate, the charger board exhibits a 25c temp rise over ambient inside my auxiliary box - not too bad. My previous high temp reading was about 55c in the trunk, so this will take it to about 80c. The limit for the ABS plastic box is closer to 100c before strange things start to happen, so it should be OK. I'm thinking it will be fine without a heatsink or a fan, though both could be added easily for use while charging.

After I draw it down again while testing wake up capability, I'll see how it behaves with a 15A charge rate. I'm not sure that I want to always pull 15A from the car's electrical system even with the engine running - that's around 180-220w.

 

[EricSan]

I'm wondering if these two cameras will STILL be enough to keep it awake when there is a higher state of charge (say, 15v) and the corresponding current draw will be lower than 0.57A...

I have some good results to post here as well! I'm running two Vantrue cameras (N4S and one channel of the S1PM) to see if their combined current draw is sufficient to keep the BMS awake. It worked when the battery was nearly depleted and now that the battery is fully charged, these two cameras also provide enough power draw to keep the BMS awake. This is the result that I was hoping for!

 

[DigitalCorpus]

Wow, a lot happens when you're gone for nearly two years. Glad you continued this. Won't make any requests because I cannot promise more than this post atm, but I'd like to refine your SoC voltage points to represent available capacity from the area under the curve. Anyhow, thanks for continued work and details you've put out there 🙂

 

[EricSan]

So here is the latest photo shoot. I rebuilt my original LTO battery pack a little bit. Originally, I built the HK3-C regulator into the pack and had just a 5v output for my Viofo A139 Pro camera. It worked great until I started testing multiple cameras at the same time and I only had one 5v output. Oops… To handle multiple cameras, I added a 12-16v output (direct from battery) next to the 5v dedicated Viofo output. But this wasn't a very neat job and my original battery held the Viofo hardwire voltage regulator hostage so I couldn't easily swap to my new larger capacity battery without having to ditch my original camera that I still like very much (and has all of the wiring stuffed above the headliner). So, I just reconfigured my original battery for a second time. Oh, how I love redoing work that I've already done twice before... I think I'm done fooling around with my original battery this time 🤣(my wife says prolly not...)

So, here it is, a little cleaned up. I flipped the rear panel inside out and installed a new 3-position speaker spring clip terminal. I added a new set of 5-way binding posts for the ground spade connectors. I covered the remaining hole with blue painter's tape. It’s not beautiful, but it looks cleaner than before. Since I discovered that the Supnova battery meter ALWAYS draws power, I rearranged its wiring so that when the BMS is turned off, the meter is disconnected too.

Given how much more capacity the newer battery has, it's really not that much larger. Here are a few side-by-side images of my original 276Wh battery (blue) and the newer 1104Wh battery (black). Batteries are now fully interchangeable and compatible with any dashcam that comes my way in the future. The speaker terminal spring clips make installing a new hardwire kit quick and easy. My newer and larger battery should provide nearly seven work days with 4 cameras running simultaneously in LBR parking mode - and likely more with some driving around. Now it will be easier to compare video footage across cameras and do some additional feature testing.

And it wouldn't be three days before Christmas without a pine needle laying on the rug. I'm sure I'll find that one again in another week or two when it stabs me in the bottom of the foot and sends me dancing and cursing across the floor...

 

[GPak]

Wow, I expected the 1104 Wh battery to be much larger!
I guess this is the result of the very compact cell arrangement designed.
Both batteries look very nice and neat !

 

[GPak]

Don't you just love it when theory matches practice?!?!? 😉

My battery died (dashcam hardwire kit shut things down at 11.6v) last night after running for 210.6 hours (nearly 9 full days). My predicted runtime was 202 hours. Running my calculations backwards means either there is more battery capacity than 1104Wh (more likely), or that I obtained an overall efficiency of 99.2% (less likely). Either way, SUPER AWESOME!!!

Here is the battery discharge curve that I created from a series of BMS app screen captures. My timing intervals were clearly not uniform, but the curve is a near perfect match to the theoretical discharge curve for LTO cells. There is one anomaly in the data at about the 88 hour mark (this bump is also present in the manufacturer's data shown below). I think this is when I was playing around with seeing if the BMS/camera would wake up and turn on after being turned off with the BMS power button. I think I inadvertently left my external PSU connected for about 5-10mins after one of the wake ups, which accounts for the little bump in the curve. I also played around with adding extra draw with my power resistors, which is likely what brought the voltage back down again. If you ignore the one or two data points that are not in line, the rest of the curve is near perfect.

I only got another 1 hour of battery time when I bypassed the hardwire kit shut down setting (11.6v) and let the BMS shut things down at the Cell UVP Under Voltage Protection level of 1.800v per cell (10.8v for the entire battery). Conclusion (for me): the hardwire kit with the low voltage protection circuit disabled is not worth the trouble, I would only gain 1 more hour for a single camera. This would translate to an extra 15-20mins when running 3 or 4 cameras.

I also noticed that the Coulomb Counter in the BMS reset the remaining battery capacity when the voltage bottomed out. It had been reading rather high as time was going by.

Looks like those "used" LTO batteries are operating darn near 100% capacity.

View attachment 88603

EDIT: Hmmm... looking at the official Toshiba SCiB documentation concerning discharge rates for this exact battery pack, it looks like that little bump in the discharge curve at about 88hrs is characteristic of the battery construction and not an artifact of my interruption of the discharge process. Check out the blue trace (0.2C) in the chart below:
View attachment 88821

Thus, this bump is not an anomaly, but a real phenomenon.
Based on this, I edited my Post#463 to correct the V-SoC dependency at the location of this bump.

... but I'd like to refine your SoC voltage points to represent available capacity from the area under the curve...

As you suggested, I calculated the V-SoC dependency using actual numerical data from the Eric's table above, calculating available capacity from the area under the curve - the most accurate method.
I don't know how to attach the Excel spreadsheet, so here's a screenshot:
(I hid the columns with calculations for better visibility of the final actual results highlighted in blue, and for comparison, I added the rounded numbers highlighted in red, for easier memorization).
The rounded numbers are within ±2.7% of the actual data.

 

[EricSan]

I calculated the V-SoC dependency using actual numerical data

AWESOME! Thank you for putting that table together, GPak!

Wow, I expected the 1104 Wh battery to be much larger!

Overall dimensions are 14" long x 7.5" wide x 8.5" tall (including my auxiliary box). Without my aux box, the LTO pack itself is only 5" tall.

I added one more item to my big LTO pack: some adhesive rubber feet to keep it from sliding around in the trunk. I picked them up from Walmart, they are a bit more than 1/4" tall and fairly flexible. They grip the top of my desk VERY well.

 

[EricSan]

After playing around with the new JK BMS for a little while, here is what I am finding about its behavior:

I set my charger board to provide 16.1v, or 2.68v per cell (approx, precision seems impossible).
The Smart Sleep voltage level can be set anywhere within the voltage range of the battery (12-16v) and it works, reliably putting the BMS to sleep. Pretty cool!

With Smart Sleep(V) set to 2.47v (14.82v), the BMS goes to sleep right where expected, and *sometimes* wakes up when a charge is applied.
Reduced Smart Sleep(V) to 2.45v (14.70v): BMS goes to sleep right where expected, and still only *sometimes* wakes up.
Reduced Smart Sleep(V) to 2.42v (14.52v): BMS goes to sleep when expected, but now it *CONSISTENTLY* wakes up when a charge is applied. Cool - this is what I was looking for!

It's kind of a bummer that the Smart Sleep wake up is a bit inconsistent within the top 20% of the battery's voltage range, but it does work, so I think I'll leave it enabled.

So, it looks like I have the following functionality:
For the top 20% of the battery State of Charge, the BMS will remain always on (just like my original, smaller LTO battery). When the SoC drops below about 14.5v, the Smart Sleep function will put the battery to sleep after one hour of inactivity and it will wake up again when the car starts. I need further testing to make sure this actually happens.

Once the SoC is below the 80% mark, the Smart Sleep function will engage, saving about 90h per week of wasted battery power (nights and weekends when it otherwise would have remained on), which amounts to about 100w of power saved each week. That's about 20h of additional runtime for a single camera, so Smart Sleep seems worthwhile. Smart Sleep will be especially useful as the snow melts and it warms up again and I ride my bike more often. During the summer months, my car might live in the garage for a week at a time without going too far.

It's time to put the new battery into service and see how it behaves in the car...

 

[GPak]

Your results are very similar to mine.
My BMS consistently wakes at a 14.64V (2.44V per cell) because I increased the charging voltage to 16.2V (2.7V per cell).
I think if you increase the charging voltage to 16.2V, your BMS will also consistently wake at about 14.64 V.

In my case, I set the smart sleep so that the battery shuts off after 5 hours, since the current consumption in parking mode is less than 0.57 A, and I need the dashcams to operate for 5 hours.
(my work schedule: 5 hours + 1 hour lunch break + 5 hours).
My battery usually operates in the range of 80% to 50%, so the smart sleep is working well for me.

 

[EricSan]

Here is my larger LTO battery in its new home. I spent the afternoon making things a bit more tidy by getting rid of the dangling wires in my car and running video and power cables above the headliner and back to the trunk. I was smarter this time and ran a string above the headliner that I left in place so pulling future wires will be easier. I should have done this the first time…

I now have three cameras hardwired and still have the ability to connect one or two more. Initial testing went well: the cameras turn on when I start the car and they transition into parking mode when I turn off the ignition. I turned off all of the cameras and the BMS went to sleep an hour later. So far, so good! 👍

I'm thinking that I might get funny looks and likely denied entry if I try to park my car in a federal building parking lot now. This many wires will look scary to someone… Maybe I'll bundle the wires in some black tape to hide the colors and keep things together a little better.

I don't know how much value this adds at this point, but here is the schematic of my new battery. Moving the 12v to 5v regulator for the hardwire kit from inside the box to outside the box simplifies things just a bit. The larger 25A charger board does indeed have an anti-backflow diode between its two positive voltage terminals, so I didn't add one to this larger battery. I specifically wired the black/ground output wire so that it connects to the P- or "input" side of the BMS so that when the BMS is turned off, there is no voltage on the output connectors. I made this change after I realized that the Supnova voltage meter ALWAYS draws power - I measured 1.95mA at 14.88v (0.03w) with the display off. Its not very much, but it is constant. As a result, I rewired my smaller 6S LTO battery to follow this same schematic (the smaller 20A charger board definitely needs an anti-backflow diode). All of the positive voltage wires between the XT-60 input connector and the battery terminals are 10g wire, all of the ground wires between the input connector and the battery are 11g wire. The automotive fuse holders are 10g wire. The 12g automotive fuse holders that I used in my smaller 6S battery generated heat that was in excess of body temp with just 10A of current running through them. I suspect the weakness in the construction of the automotive fuse holder lies in the specific junction between the wire and the metal tab that holds the fuse. My rule of thumb is that any wire that feels warm to the touch is too thin and needs to be replaced with heavier wire, hence my use of 10g fuse holders. I used an 8g automotive fuse holder with a 20A Maxi fuse for my direct to battery charging wire. The red, black, and yellow output wires to the speaker spring terminals where the hardwire kits attach are all 18g wire.

Total power draw while the BMS is powered up (and the engine is off) is just about 1.1w per hour from the BMS (0.7w), three hardwire kits (0.34w), and the Supnova volt meter (0.03w). Once the BMS goes to sleep, there shouldn't be any parasitic power consumption.

As for Bluetooth performance between the older and newer JK BMS units, I’m finding them to be very similar across my specific implementations. The older unit with a stronger Bluetooth signal is built into a box with metal panels. The newer unit with a weaker Bluetooth signal is built into a plastic box. From the inside of my trunk, the end result is that the range of each of my batteries is just about identical 🙂

 

[GPak]

......
I specifically wired the black/ground output wire so that it connects to the P- or "input" side of the BMS so that when the BMS is turned off, there is no voltage on the output connectors. I made this change after I realized that the Supnova voltage meter ALWAYS draws power - I measured 1.95mA at 14.88v (0.03w) with the display off. Its not very much, but it is constant. As a result, I rewired my smaller 6S LTO battery to follow this same schematic .....

Thanks for posting this, Eric!
I didn't think that a voltmeter could consume power when switched off.
On original 7S battery, my voltmeter was connected to the negative terminal P- of the BMS, but later I moved it to the negative terminal of the battery B-, because I liked the ability to check the voltage even when the BMS/battery was switched off.

I checked my similar voltmeter of a different brand and can confirm that it also consumes about 2 mA in the off state.
Although 2 mA is not that much, it is still parasitic consumption, so I will probably also reconnect the negative wire to the P- terminal.

 

[EricSan]

I moved it to the negative terminal of the battery B-, because I liked the ability to check the voltage even when the BMS/battery was switched off.

Originally, this was my logic as well. The convenience of a voltage level check at any time was great. Then I discovered the constant power draw. What a bummer… And now it happens with two different meters. I wonder what part of the meter’s circuit draws power? I’m envisioning a very low current “polling circuit” that watches for the button push.

 

[EricSan]

Over the past little while, I’ve noticed increasing resistance in the BMS balance wires for my smaller LTO battery. Today, it threw an error message because the resistance was too high. I checked all of the wires where I crimped them and they were fine. It seems that the problem was a result of oxidation on the balance wire harness where it connects to the BMS. Unseating and seating the harness a few times seemed to restore all of the wire resistances to about 0.250 ohms.

I also used the lithium battery load tester to drain the fully charged battery. I set the low voltage cutoff to 11.4v and set it to a constant power draw of 15w. I was happy to see the capacity come up just shy of 20Ah and the energy come up just shy of 265Wh. Had I fully depleted the batteries to 1.7v or so, it probably would have come closer to the expected total of 276Wh. At 15w of power draw, the battery lasted just more than 17.5hrs. The overall efficiency rate is somewhere in the 95% to 97% range. Pretty cool!

 

[EricSan]

Just a heads up - I just encountered another operational hiccup from my original 6S LTO battery pack. I've been playing with the charge and discharge rates to measure the battery capacity and today I started charging it again only to have it stop charging within a few mins. I wanted to charge it to it's Nominal state of charge (13.8v) and put it away for a little while, so I was keeping an eye on it in the app. After a few mins, I opened the app to see that the battery was not yet fully charged, but had stopped charging. 🤔

I popped it open and checked the fuses, but they were still good (I have one fuse at the XT-60 input and another between the charger board and the BMS. Both were good, but now I was confused. It took me a few minutes to track this down, but it appears that the Constant Charge adjustment pot had oxidized and the wiper was not making proper contact. I discovered this when I unplugged my charger and plugged it back in again and the app showed me a charge rate of over 15A (!!!) for a few seconds and then the charge rate dropped to 0A. Needless to say, 15A was maxing out my Meanwell PSU...

Fortunately, I had set the "Continued Charge Curr.(A)" to a 12A limit when I was having trouble with the battery backflowing into my car's start battery (before I added the diode) - so the BMS was shutting things down, just like it should. After spinning the adjustment screw on the charger board back and forth a few times, things were good again. No lasting trouble, but I was surprised to see that the adjustment pot was causing issues.

In more than 15 years of using these multi-turn pots, I've never had one of them drift or just go "open circuit" on me before. I suppose this is why some people use these kinds of pots only for performing adjustment, then replace the pot with an equivalent value resistor...

 

[GPak]

These adjustable potentiometers are not very precise to adjust, but after setting them, I have never had to re-adjust them due to poor contact.
It's good that we have a redundant protection system, so even if the charger's potentiometer failed or drifted out of initial adjustment, the BMS kicked in and stopped the charging.

The most valuable thing in all of this is that you assembled this battery yourself, and accordingly, you diagnosed the problem yourself and promptly fixed it.
Just imagine what it would be like to discuss such a problem with the manufacturer's customer support located on the other side of the world.