LifePO4 GBS Amp Hour Testing 2.5v to 3.6v per cell

[createthis]

Let me give you an idea how my brain works after a thread like this. Anytime I experience a lot of backlash, I attempt to learn from it.

When I released the 4/0 cable crimping video I immediately started trying to figure out how to perform a tensile test of the crimped 4/0 cable on camera. So far, I'm drawing a blank because I read the breaking strength of 4/0 cable is somewhere in the range of 8000 - 10000 lbs. I think I'll need to weld up my own rig using 10 - 20 ton hydraulic jacks to do the job. Not impossible, but not trivial either.

I'd *love* to do a video comparing the crimps made by a professional crimper and the cheap TMS crimper. Nothing would make me happier. However, I need a way to test the crimps first, and second I need to gain access to one of those professional crimpers.

Second, from this thread and the thermal concerns raised, I immediately started looking around for a thermal camera I could afford, beg, or borrow. I think it would be really awesome to see the thermal issues on camera. Then I can easily make a statement on camera like, "don't do this unless you want to burn your house down". Probably, I and everyone watching the video will learn a lot.

These videos are not self contained. They're a journey. I'm not trying to screw anyone. I'm here to learn.

[SunEagle]

Let me give you an idea how my brain works after a thread like this. Anytime I experience a lot of backlash, I attempt to learn from it.

When I released the 4/0 cable crimping video I immediately started trying to figure out how to perform a tensile test of the crimped 4/0 cable on camera. So far, I'm drawing a blank because I read the breaking strength of 4/0 cable is somewhere in the range of 8000 - 10000 lbs. I think I'll need to weld up my own rig using 10 - 20 ton hydraulic jacks to do the job. Not impossible, but not trivial either.

I'd *love* to do a video comparing the crimps made by a professional crimper and the cheap TMS crimper. Nothing would make me happier. However, I need a way to test the crimps first, and second I need to gain access to one of those professional crimpers.

Second, from this thread and the thermal concerns raised, I immediately started looking around for a thermal camera I could afford, beg, or borrow. I think it would be really awesome to see the thermal issues on camera. Then I can easily make a statement on camera like, "don't do this unless you want to burn your house down". Probably, I and everyone watching the video will learn a lot.

These videos are not self contained. They're a journey. I'm not trying to screw anyone. I'm here to learn.

Surprising the thermal cameras have come way down in cost to the ones I was trained on back in 2010. The one I used was made by FLIR and it cost about $8000. Now you can get one similar in accuracy (80 x 60 pixels) called an E4 for under $1000 and one that is good enough that attaches to your smart phone called a FLIR One for about $250.

[createthis]

Surprising the thermal cameras have come way down in cost to the ones I was trained on back in 2010. The one I used was made by FLIR and it cost about $8000. Now you can get one similar in accuracy (80 x 60 pixels) called an E4 for under $1000 and one that is good enough that attaches to your smart phone called a FLIR One for about $250.

Yeah. I'm currently considering the Therm-App Hz. $1600, 384 x 288 resolution, interchangeable lenses. Seems like it would be a good choice for youtube videos since the refresh rate is high. Need to talk to some people to get a feel for how accurate it is first. I don't think I need incredible accuracy. I want to be able to communicate. I can always use a regular laser thermometer if I need accuracy. Requires an android phone, but I need a new phone anyway. Still using an old iPhone 4s.

Those FLIR E4s are (or were) hackable to be an E8 BTW. They all include 320x240 sensors, but they're software limited, much like Rigol's oscilloscopes. My research indicates the E4 - E8 range comes with a pretty crappy lens though. You learn pretty quickly in photography that it's all about the lens.

[SunEagle]

Yeah. I'm currently considering the Therm-App Hz. $1600, 384 x 288 resolution, interchangeable lenses. Seems like it would be a good choice for youtube videos since the refresh rate is high. Need to talk to some people to get a feel for how accurate it is first. I don't think I need incredible accuracy. I want to be able to communicate. I can always use a regular laser thermometer if I need accuracy. Requires an android phone, but I need a new phone anyway. Still using an old iPhone 4s.

Those FLIR E4s are (or were) hackable to be an E8 BTW. They all include 320x240 sensors, but they're software limited, much like Rigol's oscilloscopes. My research indicates the E4 - E8 range comes with a pretty crappy lens though. You learn pretty quickly in photography that it's all about the lens.

Yep. One of the main things about thermography was being able to get a picture in focus. If it was blurry then there was no way you would get an accurate temperature reading. But then again it was the software that dug out the detail once the picture was taken.

[Sunking]

What I don't appreciate is the way you're all going about it. Threatening and making assumptions is counter productive. You all could have taken an entirely different tone with the conversation by asking questions and pointing out risks and recommending ways to mitigate those risks..

We tried that many threads ago and gave up. We all bent over backwards trying to help you and show you the ways. All you want to do is argue and cry foul. I even warned you this could happen and you did not listen.

Don't waste your money on a Thermal Camera as it is not going to show you much of anything. If every connection is making good contact, there is no heat. But that still misses the point. There is no good reason you can come up with to be using 4/0 AWG, and now that you did are really usingg the wrong hardware compounding the error. If you are using 4/0 AWG with Flexible cable strands you had better use long barrels and 2-Hole Terminals made for 3/8-inch hardware connected to a Term Plate or Buss.

Here is the point. With DIY Solar, there is no reason you should exceed 100-amps on any portion of the circuit. At 100-amp limit means up to 5000 watt panels on 48 volt battery with a 5000 watt Inverter. More than enough for almost every application you can think of. With those limits come limits on equipment.

To start with a 100 amp limit means no conductors larger than #4 AWG copper. Secondly Mr/Ms DIY can actually afford to buy the Hand Crimpers and handle the wire using Standard Class B Stranding. Real easy to strip, terminate, and work with. Although I would never advocate using 1-Hole Terminals on currents above 20 amps, 100-Amps is the standard practice limit at 100 Amps. Why you ask?

1. It takes a lot of surface area contact are to transfer and conduct 100 amps or more. Requires proper torque and periodic checks

2. Large Cables and Lugs and large and heavy. which will add quite a bit of stress to the point of attachment, that can easily break free. Show me any 1-Hole terminal and I can take it apart with no tools.

So with a 100 Amp Limit means you have some limitations. Dpending on battery voltage determines maximum charge and discharge limits in watts.

12 volt @ 1000 watts
24 volts @ 2000 watts
48 volts up to 5000 watts.

Largest Charge Controller you can buy is 80 amps, well Midnite Solar Classic is 96 amps. Ever wonder why? 80 amps on a 12 volt battery is a 1000 watt panel.
Based solely on your batteries maximum acceptable operating limits (0.5 Volt Sag or 4% loss) can go up to 50 amp or 600 watt panel and Inverter using 6 AWG minimum wire. Note 4% is about twice what a Solar System should be on batteries. The .5 volt drop is battery only. Cable loss is on top of that. Otherwise 25 amps at 2% drop.

You said you learned your lesson about battery Internal Resistance. If that were true, you would understand 150 amps is silly. Who would design a system with 12% loss, and cannot use the battery capacity. That is what you built. You put the cart in front of the horse. If you had really needed 12 volts @ 150 amps would have required a different battery. One that would last as long and most likely longer, and cost 1/4 of what you paid like an Odyssey PC2250 or other like 12 volt @ 110 AH AGM. Note the published Ri is .002 Ohm's and at 150 amps is .3 volts or 2.5%.

[createthis]

You said you learned your lesson about battery Internal Resistance. If that were true, you would understand 150 amps is silly. Who would design a system with 12% loss, and cannot use the battery capacity. That is what you built. You put the cart in front of the horse. If you had really needed 12 volts @ 150 amps would have required a different battery. One that would last as long and most likely longer, and cost 1/4 of what you paid like an Odyssey PC2250 or other like 12 volt @ 110 AH AGM. Note the published Ri is .002 Ohm's and at 150 amps is .3 volts or 2.5%.

Sunking, I think you're angry that Xantrex made the SW2000 inverter, not that I used it in a system.

Who would design a system with 12% loss, and cannot use the battery capacity. That is what you built.

I'll take a moment to help you refactor that statement so it's more helpful and less abrasive:

"You designed a system with 12% loss that cannot use the available battery capacity. I wouldn't have done that and other electrical engineers wouldn't have done that because there is a better way that is more efficient."

[karrak]

You stated in your video that the voltage dropped from around 13.1 volts to 12.5 volts or a voltage drop of 0.6V with a current load of 146A, contrary to what Sunking is saying even if you doubled this voltage drop to 1.2 volts you would still be able to use most of your battery capacity. A no load battery voltage of 12 volts for LFP batteries equates to less than 10%SOC. If we subtract the 1.2 volt drop from 12 volts we get 10.8V which is above the 10.5 volt limit that most inverters will cut out at.

Battery resistance is not a fixed, easy to calculate figure. If it were fixed, the voltage would drop suddenly when you put a large load on the battery and then stay constant at the lower voltage until you remove the load. What you see is the voltage drops when you apply the load and then keeps dropping after this. This means that the internal resistance is increasing the longer the load is applied. This is especially true and one of the major drawbacks with lead acid batteries.

I think the best way to check what the losses in the battery are is by measuring the voltage just before you apply the load with the battery SOC between 75% and 25%, then after you apply the load and then say two minutes after you apply the load. Subtract the voltage under load from the noload voltage, then multiply this by the current. this gives you the losses in the battery. So in your case your losses are (13.1-12.5)*146 = 87.6 Watts, P_{BatteryLosses}= (V_NoLoad - V_load)*I_load

To calculate the losses in the cabling, fuses, shunts, connectors etc apply the load, measure the voltage at the battery terminals and at the input to the inverter. Subtract the voltage at the inverter terminals from the battery terminal voltage and multiply it by the current to get the power losses. P_CableLosses= (V_inverter - V_battery) * I_load

Simon

[karrak]

One that would last as long and most likely longer, and cost 1/4 of what you paid like an Odyssey PC2250 or other like 12 volt @ 110 AH AGM. Note the published Ri is .002 Ohm's and at 150 amps is .3 volts or 2.5%.

For some reason trying to add a comment to a post is coming up with the error "No Text" so here is my comment

The Odyssey PC2250 is a heavy duty battery designed to start motors and other intermittent loads. It is totally unsuitable as a deep cycle battery.

From the Odyessy website
"Typical deep-cycle life at 25°C/77°F at a 5-hour rate 400 cycles at 80% DOD"
So at a current draw of around 20A we only get 400 cycles, compare this to the 100Ah LFP battery that createthis is using which will give 2,000-3,000 cycles at 80% DOD.

Both batteries cost around the same, again from the Odyessy website the cost of the PC2250 is $520.

[inetdog]

For some reason trying to add a comment to a post is coming up with the error "No Text" so here is my comment

Any text that you add inside the QUOTE tags of the other user's text on a Reply post does not count against the minimum. Be sure you are typing after end of the automatically quoted text.
Comments, on the other hand, are just weird and I suggest not using them.

[karrak]

You said you learned your lesson about battery Internal Resistance. If that were true, you would understand 150 amps is silly. Who would design a system with 12% loss, and cannot use the battery capacity.

You would do it if it is cost effective. In Jessie's case he will probably only want to run his microwave for a few minutes for a few times a day and the % of the power that he will draw from the battery at these power levels will be less than 10% of the total power cycled through the battery.

If I were running high currents for any length of time I would put a temperature sensor wedged between the middle batteries and/or thermally bonded to one of the battery terminals to make sure that the batteries are not getting too hot, could be a wireless one.

One thing that concerns me about Jessie's setup is that the inverter and solar controller which can generate more heat than the battery are packaged with the battery and thus will heat the battery up. Temperature is one of the largest factors that reduce battery lifespan, this is true of not only LFP batteries but also lead acid batteries. If this is a test system it probably doesn't matter, but something to consider.

Simon

[PNjunction]

Question: Is the Midnight Classic and inverter rated for horizontal use as seen in your case?

I'm a little wary of the horizontal wooden mounting plate. I busted a kid's design chops about using balsa wood with his 18650 project enclosure, so to be fair ...

If you've got the Class-T fuse, use it!

Reason is it appears you've got a little bit of a spaghetti-nest going on underneath, and if one of your crimps overheats, burns or drops the wire *prior* to any breakers, we've got a problem.

Thermal camera won't be much good when the unit is closed too for seeing underneath wiring / meter connector issues. Multiple openings/closing and wires bunching a bit for testing - if that side-torques your connections just a bit loose, now you've got a hot-spot connections you can't see, covered by wood overhead.

The reason I'm pointing this out is that the unit for all practical purposes won't be opened up like it would be for spot-testing the stuff underneath.

[SunEagle]

Question: Is the Midnight Classic and inverter rated for horizontal use as seen in your case?

I'm a little wary of the horizontal wooden mounting plate. I busted a kid's design chops about using balsa wood with his 18650 project enclosure, so to be fair ...

If you've got the Class-T fuse, use it!

Reason is it appears you've got a little bit of a spaghetti-nest going on underneath, and if one of your crimps overheats, burns or drops the wire *prior* to any breakers, we've got a problem.

Thermal camera won't be much good when the unit is closed too for seeing underneath wiring / meter connector issues. Multiple openings/closing and wires bunching a bit for testing - if that side-torques your connections just a bit loose, now you've got a hot-spot connections you can't see, covered by wood overhead.

The reason I'm pointing this out is that the unit for all practical purposes won't be opened up like it would be for spot-testing the stuff underneath.

Correct. A thermal camera must be able to be focused on a spot the emits thermal energy. That energy can be blocked by other material or if the wrong emissivity setting is used on the camera you will get an incorrect temperature reading. Best to have the camera look at the insulation of a wire close to a connection point and not the metal part of the wire or connector.

[createthis]

Question: Is the Midnight Classic and inverter rated for horizontal use as seen in your case?

The midnite classic is indeed rated for vertical mounting. It is mounted vertically in the case. The control panel is separate via an RJ11 cable.

The inverter cannot be mounted vertically, per the manual, so it is mounted horizontally. This is again how the manual recommends.

If you've got the Class-T fuse, use it!

What's wrong with the thermal DC breaker? We don't use fuses in AC panel anymore. Why should I use them in a DC panel if there is a reasonable DC breaker available?

Reason is it appears you've got a little bit of a spaghetti-nest going on underneath, and if one of your crimps overheats, burns or drops the wire *prior* to any breakers, we've got a problem.

The breaker is the first thing in the circuit. The cable length is maybe 5 inches long. It was as short as I could make it. A class-t fuse would have had to go in exactly the same place.

I think most of the spaghetti nest you see is the RJ11 cables from the various control interfaces. I need to shorten them. I made the 4/0 cable runs as efficient as I could. I spent days moving things around and trying out different configurations. It's possible someone else could do a better job. A larger case and 48v would be the best solution.

Thermal camera won't be much good when the unit is closed too for seeing underneath wiring / meter connector issues. Multiple openings/closing and wires bunching a bit for testing - if that side-torques your connections just a bit loose, now you've got a hot-spot connections you can't see, covered by wood overhead.

I'm aware of the limitations of thermal cameras.

The reason I'm pointing this out is that the unit for all practical purposes won't be opened up like it would be for spot-testing the stuff underneath.

Yup.

[createthis]

I'm hearing three things here:

1. The mods all seem to agree this is a "battery bomb".
2. One reason why it is a battery bomb is I2R heating of the battery and cables during normal high amp operation.
3. The second reason why it is a battery bomb is because I didn't use power lugs (the double long ones) and a certified crimper and I may not have torqued my connections properly.

I'm curious about #3 and I'd like to devise a series of tests to determine whether this should really be a concern or not. I've had crimps fail before. Both mine and commercially purchased wiring. I understand the risk here. That doesn't change the fact that people other than me ARE using these lugs and crimp tools. I'm not the first person to buy these products. I'd rather be able to prove on camera that it is or isn't a problem than wonder about it. I don't think sticking my head in the sand is helpful.

I was aware of #2 before I started building the machine. I'm not very worried about it because how I use the machine is directly proportional to the amount of heat generated. Yes, people make mistakes sometimes. I'm aware of the risk involved. I'm also happy to make another video talking about it/proving it.

[SunEagle]

I'm hearing three things here:

1. The mods all seem to agree this is a "battery bomb".
2. One reason why it is a battery bomb is I2R heating of the battery and cables during normal high amp operation.
3. The second reason why it is a battery bomb is because I didn't use power lugs (the double long ones) and a certified crimper and I may not have torqued my connections properly.

I'm curious about #3 and I'd like to devise a series of tests to determine whether this should really be a concern or not. I've had crimps fail before. Both mine and commercially purchased wiring. I understand the risk here. That doesn't change the fact that people other than me ARE using these lugs and crimp tools. I'm not the first person to buy these products. I'd rather be able to prove on camera that it is or isn't a problem than wonder about it. I don't think sticking my head in the sand is helpful.

I was aware of #2 before I started building the machine. I'm not very worried about it because how I use the machine is directly proportional to the amount of heat generated. Yes, people make mistakes sometimes. I'm aware of the risk involved. I'm also happy to make another video talking about it/proving it.

Yeah about doing something for a while without bad results. I have heard of people playing Russian roulette in Asia many times and surviving. Yet it is still pretty dangerous and will eventually have fatal results.

When a product is made in the US it goes through a battery (excuse the pun) of failure tests. It has to first not fail and then if it does fail it can't hurt the user. IMO not getting shot 49 out of 50 times would not be classified as being safe.