DIY portable, useful LiFePO4

[Luv2Go]

BTDT. Booster has good advice. My ancient Weller WTCPT (60W) does an OK job if I use the biggest tip I have for it, however one of these days I really need to buy a soldering gun, at least 150-200W, maybe like a D650. The advantage of the gun over the iron you have is you control the heat with the trigger. And definitely helpful is 60-40 rosin core with a tin of rosin flux to clean the tip.

[booster]

Originally Posted by markopolo

Thanks for the info. Yes, the 120W is BIG. It is a Weller with no controls, just plug it in. Huge 1/2" chisel type tinned tip. I did tin it and wipe it on the damp sponge. Yes, I was getting charring & had to clean it up. I'll fine tune my methods based on the advice received & try to lay down some big puddles of solder. The plan is for 6 X 10 AWG, pos & neg so 12 solder connections needed.

I could have used clips for the test but I'm glad I tried soldering on those pads.

That sounds similar to the huge Hakko that I have. If I let it warm up to it's steady state temperature, I bet it gets to close to 800*F which is way too hot for flux.



Having silver in the solder makes mostly for higher strength, I think, and you would have to check the melting point. One thing when you look at listed melting temps, they should have to solidus and liquidous. For things like electrical connections you usually would want the two very close together so it runs right away and goes very liquid to wick. 60/40 will be as much as 100*F lower melt than the 97/3 IIRC and probably also more flexible once in use, which is necessary in electrical connections that have things moving like wires or from heat.

[booster]

Originally Posted by Luv2Go

BTDT. Booster has good advice. My ancient Weller WTCPT (60W) does an OK job if I use the biggest tip I have for it, however one of these days I really need to buy a soldering gun, at least 150-200W, maybe like a D650. The advantage of the gun over the iron you have is you control the heat with the trigger. And definitely helpful is 60-40 rosin core with a tin of rosin flux to clean the tip.

We had those Wellers as bench irons for a long time, but used the one piece temp controlled the rest of the time as were soldering in place while building machinery. My old Weller 60watt portable temp controlled starting tripping the garage GFCI breakers, so I got a new one. Same problem. The portable Wellers use a thermal switch internally, that basically slides, so it must arc too much and trip the outlet. The 60w Hakko in the link is sensed electrical control and transfers heat very quickly so smaller tips have more "apparent" heat capacity and bigger tips will do some quite large stuff for a 60W iron.


Personally, I never liked the guns as they get too hot, too quickly, and have very low heat capacity so really tough to be save on critical stuff like PC board work where a little extra temp can damage components or lift the foil. Best for me is a iron that is temp controlled with a big enough tip for the heat capacity needed, and short reaction time so it is basically impossible to overheat anything if set correctly.

[Luv2Go]

Originally Posted by booster

... The 60w Hakko in the link is sensed electrical control and transfers heat very quickly so smaller tips have more "apparent" heat capacity and bigger tips will do some quite large stuff for a 60W iron.
....

Yeah, I should upgrade my soldering iron to the Hakko. Probably wouldn't need the gun. Most of my need for the gun was to solder lugs on 1/0 and larger wire or to join large wires together without lugs, nowadays I use lugs/splices and a hydraulic crimper which makes a much better connection.

[Winston]

Originally Posted by Luv2Go

Most of my need for the gun was to solder lugs on 1/0 and larger wire or to join large wires together without lugs . . .

Just put a couple of copper lugs on some 2/0 wire - - used large diameter 60/40 rosen core solder and a propane torch!

[markopolo]

The BMS app issues were phone related for me. The very newest version as of today ( 3.1.1015 ) runs on an older phone I have no problem. I think there are 52 settings that can be changed! Temperature included.

The version in the Play Store is locked down it appears. It's for viewing only.

Some screenshots from version 3.1.1015 ( 8 x 1.5V AA alkaline in series just for a quick test to get the BMS to start up )

new version 01.png

new version 02.png

new version 03.png

new version 04.png

new version 05.png

new version 06.png

new version 07.png

new version 08.png

new version 09.png

new version 10.png

new version 11.png

[markopolo]

Correction: Previously I said: "The plan is for 6 X 10 AWG, pos & neg so 12 solder connections needed."

That was a mistake. I never intended to solder to the (probably rare) B+ (Pos) pad. That pad was a "customer design for a car start application" per the vendor. It isn't needed. What I should have typed is: The plan is for 6 X 10 AWG, neg input & output so 12 solder connections on the BMS needed.

Soldering update: I think the 120W iron will do the job well after a bit of soldering today. I just had to apply heat longer. The 80W iron might have a faulty thermostat as it doesn't seem to get hot enough even at it's max setting. Power consumption is intermittently 30W even when adjusting it from 200C to 480C. I did see 80W for about 1 second when first plugging it in.

BMS PC software: Installed & easy to use. It can be set to log to Excel. It gives access to every single setting I think. I saved a copy of the current eeprom settings so I can always reload the original settings.


JBDTools.jpg

DC to DC Charge controller PC software: Installed & easy to use. You can create 10 memory presets which are stored on the device itself. They can be recalled on the panel installed on the charge controller box or by PC program.


DPS5020.jpg


DPS5020 storing presets.jpg

[markopolo]

The 80W iron can be calibrated but I don't have a suitable thermometer.

Sn97Cu03 solder has a melting point of 227C. The 80W iron set to 228C would not melt the Sn97Cu03 solder. I calibrated the iron so when set to 228C it would melt the Sn97Cu03 solder (not instantly).

The calibration instructions stated that calibrating at 320C would ensure good linearity across the temperature control range.

Lead has a melting point of 328C. I calibrated the iron (more increases) so when set to 328C it would begin to melt lead if held for a bit.

Hopefully the 80W iron has more accurate control now.

I don't how good this method of calibrating is but it's the best I could come up with short of buying a specific thermometer. The watt meter readings make more sense to me now. Power is constantly fluctuating as expected but, at higher settings, it never drops down to 1W or less as it did before.

[booster]

Originally Posted by markopolo

The 80W iron can be calibrated but I don't have a suitable thermometer.

Sn97Cu03 solder has a melting point of 227C. The 80W iron set to 228C would not melt the Sn97Cu03 solder. I calibrated the iron so when set to 228C it would melt the Sn97Cu03 solder (not instantly).

The calibration instructions stated that calibrating at 320C would ensure good linearity across the temperature control range.

Lead has a melting point of 328C. I calibrated the iron (more increases) so when set to 328C it would begin to melt lead if held for a bit.

Hopefully the 80W iron has more accurate control now.

I don't how good this method of calibrating is but it's the best I could come up with short of buying a specific thermometer. The watt meter readings make more sense to me now. Power is constantly fluctuating as expected but, at higher settings, it never drops down to 1W or less as it did before.

Soldering iron temps are a difficult thing to do in general, as there are so many outside variables that mess up the logic behind it.


The biggest issue for most irons is how fast they can react to the temp of the tip dropping when it gets cooled from heating the objects and the solder. If it doesn't react very quickly, which is very common in non temp controlled irons, you will wind up heating the area longer and have increased risk of burning or drying out flux, oxidation of the parts, or damaging the wire insulation of sensitive parts. Going to larger tips helps the situation if they happen to match the heat capacity of what you are soldering, but make it worse if the aren't big enough.


Since the melting of the solder isn't instant at a given temp, but is more over a range of temp to fully liquid, you melt off a bit and not have it run and wick correctly if it is a bit low on temp. In some solders, especially the higher temp alloys like silver solder in the 1100*+ range you get the alloy components separating at temps not liquid enough to run. Not as often at low temp, stuff but can happen.


If you turn up the power of the iron on an adjustable but not temp controlled, you can improve the response time but have to make sure you get onto the job at the right time before the tip gets too hot and burns the flux or damages things.


The temp controlled irons are normally way more wattage than you need for the job, as they can't get too hot, so they can react quickly when quenched. The quality of them varies widely, and the only two I have actually tested that worked well were the Weller and Hakko models, with the Hakko my favorite by quite a bit. I don't recall ever seeing temp control in anything bigger than 60 watts, but they may exist. That makes it so much more skill is needed for the big item soldering applications. The big irons with a temp rating or scale if adjustable are normally giving the static full temp the iron will get to while not in use and if you set to that matching the solder, they will be too cold when actually soldering as they can't react fast enough. It would be interesting to see the model number of Marko's adjustable 80 watt iron as it may be different than I have seen in the past if it is truly dropping power to the tip to regulate temp.


You can fairly accurately check the tip of the larger irons with an IR temp gun if it will read when very close and has a small sensing spot, but best is an IR camera, I think. They are getting pretty inexpensive now, I think, but used to be very expensive. I think the first one we got while I was working was about $1000+ 30 years ago. We used them to check part temps coming out ovens, laminating presses, etc in the circuit board making business. Maybe there is a phone app by now, even. I have seen them used in getting setup parameters like heating up parts to grow them for heat shrink assembly of metal parts so they could watch the quenching happen in real time to choose best temps.


On edit, I just went out and looked at the setting on the Hakko 60 watt. It is at 350C, which is about 660F. It is very quick in and out at that temp and doesn't burn the flux. I can turn it down but it get slower. If am on a pc board doing a component replacement with a tiny tip, I would also turn it down to prevent delamination of the copper foil pads from the fiberglass board and also work with a puddle of flux to prevent hot spots.

[markopolo]

It's an inexpensive iron, around $22 on Amazon when I bought it. No brand, model 936H. Wattage is constantly fluctuating and I could not discern a difference in the randomness of the fluctuations when applying the tip to a large piece of aluminum. I have a good variety of 900M tips for it. I currently have a fairly large tip on it, 4mm diameter on a angle, makes for a flat oval footprint. The length of the oval is 3/16. I'll probably try to measure with an IR temp gun later today.

I should buy or should have bought the Hakko you mentioned or something similar.

[skunberg]

Originally Posted by markopolo

Thanks for the advice. I had tinned the wire ends and was trying put a bit of tin on the copper pad. It just wouldn't stick & I don't give up easily. The solder that eventually stuck was 97% Tin 3% Copper. No doubt you're right about the heat. The 97Sn/3Cu solder does have a higher melting point than the other solder I tried. It's also 1/8" solder so the larger quantity = more thermal mass and thermal transfer in its liquid state.

Do you think the 97Sn/3Cu is adequate or is some silver content preferred/advised?

My favorite is Sn62 Pb36 Ag02 Kester 44 series. A little silver makes things wet easier. Try it and you won't go back to 60/40.

[markopolo]

I decided to use the EDB-M05 today ( http://www.classbforum.com/forums/f2...tml#post103729 ). It is easy to setup and use the PC software.

It's a 30W tester. If testing a near full LiFePO4 pack then 30W/13.3V = 2.25A would be the maximum current. That's something like what a Maxx Fan could draw.

No lithium batteries yet but a quick series of tests on a small 12V mower battery showed that it works.

2A current used for the test. The battery had been on a float charger but the screen shot was probably the 4th or 5th quick test. The battery did recover to 12.8V after a while. These were quick tests, not completed.

It was very interesting to see the voltage sag and recover with the 2A load.

ebtest.jpg

[booster]

Originally Posted by markopolo

I decided to use the EDB-M05 today ( www.classbforum.com/forums/f23/diy-portable-useful-lifepo4-10057-2.html#post103729 ). It is easy to setup and use the PC software.

It's a 30W tester. If testing a near full LiFePO4 pack then 30W/13.3V = 2.25A would be the maximum current. That's something like what a Maxx Fan could draw.

No lithium batteries yet but a quick series of tests on a small 12V mower battery showed that it works.

2A current used for the test. The battery had been on a float charger but the screen shot was probably the 4th or 5th quick test. The battery did recover to 12.8V after a while. These were quick tests, not completed.

It was very interesting to see the voltage sag and recover with the 2A load.

Attachment 8639

That voltage sag is certainly interesting. Was this a wet or AGM battery?


We see sag in most all our higher draw stuff on AGM and usually it also recovers, but much faster than the 1 minute that your's shows. I do think that when we had wet cells the sag was bigger and lasted longer, but I am not certain. It will be interesting to see if it also happens on lithium.

[markopolo]

Sealed wet cell starter battery for a lawn tractor, a few years old.

[markopolo]

Single LiFePO4 cell fully charged would be around 3.325V (with a load).

30W/3.325V = 9A Approx 6Ah cell I think that would be a 1.5C load test.

[booster]

Originally Posted by markopolo

Single LiFePO4 cell fully charged would be around 3.325V (with a load).

30W/3.325V = 9A Approx 6Ah cell I think that would be a 1.5C load test.

Do those cells list a chart or graph of max time at various discharges, or such?



We seem to be seeing more and more low discharge and charge limits coming from the lithium suppliers, so the original claims of huge charge and discharge rates may be fading out to some degree. Of course on a monster bank of 800ah, .5C is 400 amps and above what can be supplied by anything other than Volta, I think, and I doubt anyone has enough stuff to run to use that much on inverter in a B van.

[markopolo]

No real info that I can find. They're not going to be Grade A. They'll have higher resistance for example.

They're advertised as 7Ah but the best that purchasers have reported is 6Ah. Some report worse, as bad as 5Ah but the current draw on those tests is as high as 20A from one cell! With 40 cells, 2A from each cell would likely power my smaller 600W/900W microwave oven for example. 80A*13.3V = 1,064W


Note: EDB-M05 maxes out a 30W. You choose the current though so you can test a single cell at much less than 1.5C. Example: 3.325V*1.8A = 6W , 1.8A/6A = 0.3C test

[markopolo]

Low temperature protection on the BMS does appear to work. I set the charge and discharge low temperature cut offs above freezing and put the temperature probe in between two ice packs.

I'm fairly certain that an actual load and an actual charging voltage were required for the associated protection to kick in. I first tested it without a load and not charging and nothing happened that I noticed. When tested with a load and when charging it worked.

The temperature and voltage captured in the screenshot was when I got the screen capture and not at the exact time of the charging voltage or temperature trigger event.

Low temperature 1.jpg

Low temperature 2.jpg

Low temperature 3.jpg

[markopolo]

I discovered an anomaly of this BMS - http://www.classbforum.com/forums/f2...tml#post103473 - and a better way to connect the balance wires (I think).

These boards typically have a JST type connector for the balance wires. The total amount of connector pins usually exceeds the number of batteries in series. This 4S BMS has a 5 pin JST connector and came supplied with a 5 pin wire harness. Pin 0 goes to ground, pin 1 at 3.2V, pin 2 at 6.4V, pin 3 at 9.6V and pin 4 12.8V. The last pin is a red positive wire at total pack voltage. Those were the instructions given to me and the parts supplied and usually that is the correct way to do it.

I discovered that the Pin 0 wire (ground) carried load current in parallel to the B- pad because pin 0 and B- are connected within the PCB. I was using a led bulb as the test load so it wasn't a problem. A large load would have smoked that small gauge Pin 0 wire. Cutting the pin 0 wire solved that issue. Now load current has to pass through the correct path and the BMS works as it should.

The better way IMO is to use the three JST 2 pin PH connector sockets that are on the other side of the board and also supply the unique B+ pad with total pack voltage directly from the battery pack. That would be instead of using the supplied 5 pin connector. B- is ground, the first JST 2 pin PH connector would be 3.2V, 2nd is 6.4V, 3rd is 9.6V and the B+ pad is 12.8V. Each 2 pin connector is really 1 connector as the pins are joined within the PCB. 2 pin connector #1, 2 & 3 connect directly to pin #1, 2 & 3 at the previously mentioned 5 pin connector. Pin 5 from the 5 pin connector connects directly to the B+ pad.

I think it is better electrically because, in addition to not paralleling load and charging current from the B- pad on the tiny pin 0 ground wire, the balancing wire capacity is at least doubled by using the 2 pin connectors. I ordered some JST 2 pin harnesses and will do it this way.

The important thing IMO is to not use pin 0 of the 5 pin connector on this particular BMS as it will carry load and charging current in parallel to the correct path. I think the 5 pin connector on this BMS was likely intended to be used only for testing and programming.

[markopolo]

I found a photo of the BMS I have used on a production battery.

same BMS and not using 5 pin connector.jpg

I like the way it's done & nice to see confirmation that using the 2 pin connectors is the way to go for the balance wires. The 5 pin connector is not used in this example.

No soldered wires used to carry load & charge power. It's really tempting to change my plan. I can fit more batteries into the case I have with the batteries vertical rather than horizontal though.