@davee Just some additional info, you will like part of it as it explains why you did not understand something, I will get to that in a bit.
The first 2 pictures show a Waveshare Pico UPS-B. Notice how they selected a battery that fits nicely between the header pins and is shorter than the board so it tucks up nicely under the board. I am unsure if you can see it but the battery is one continuous set of wires terminated in a JST connector that mates to the Pico. NOTE this is a different JST connector from the next item.
The last two pictures are a LILYGO T_Display_S3 V1.2, the successor to the OP's ESP32 V1.1 but other than a newer ESP32 it is the same. Here is where you got confused but by dumb luck I knew since I bought the board. The d*m JST connector on the board is incredibly tiny and for the longest time I was ordering the wrong parts. I must have at least 6 different tiny JST male connectors. I finally found a fantastic resource which I will share here for everyone to benefit JST Cheat Sheet.
Because the tiny connector is so hard to find LILYGO provide one with the bare wires. Yes, I had to cut the battery wires and solder them together because I was never able to find a properly shaped (max 20mm x 50mm) LiPo pack with the same tiny JST male plug. The last two pictures show the tiy female connector, with the supplied bare wire to tiny JST on the bottom, and the stock too big JST on the top.
NOTE I need to correct myself, the LiPo's I am looking at right now do NOT have chemistry marked on them. Each one does have 3.7V which means LiPo BUT not 100% as LiIon can be labelled 3.6V or 3.7V. BUT AFAIK all bag batteries are LiPo.
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
Hi Ron @zander,
Thanks for your contributions, although I fear this is getting some way from Display Issues. Obviously you have more first-hand experience of the actual items, as I am only looking at information on the web. Your experience and findings seem to be in general agreement with my analysis.
---------
Please note that Li-Po is short for lithium-ion polymer. The 'polymer' addition refers to modifications of the electrolyte to convert it into a semi-solid gel instead of a liquid. The electrolyte plays a crucial part in a cell, because it allows transfer of the ions needed to create a current (and to transfer the ions back to their 'starting' position during charging). The electrolyte is analogous to a road for ions, so that a more efficient transport arrangement implies the ability to handle higher currents during charging and discharging. However, the principal chemistry of the cell refers to what happens at the anode and cathode electrodes. Changing what these electrodes are made of, changes the voltage the cell produces, and consequentially what voltages are required for charging.
So far as I am aware, any of the different electrode types could adopt a 'polymer' style of cell, although obviously the electrolyte polymer must be carefully chosen to be compatible with the specific electrode types, which might present a difficult practical challenge in some cases. In addition, commercial pressures will inevitably mean that some combinations will be cheaper and easier to obtain than others.
----------
18650, as you know, is derived from the cylindrical size in mm (18mm diam, 65mm high), albeit cells on sale sometimes vary by a couple of mm or so. It does not, in any way, define the chemistry, or the corresponding voltage characteristics. I suspect the majority commonly available will be cobalt oxide varieties of lithium-ion, plus a significant showing of LiFePO4 as you suggest, but I think other chemistries will become more common.
----------
Hence, while I accept that certain cell types are more likely to purchased, and that in many cases, by chance, they will be fine, I do not accept this is a universal rule.
For example, I noted you showed a 500mAH cell, which I suspect is the smallest cell in that range that would be suitable, with maximum charge rate of 500mA. Was this by chance, or did you instinctively know the board charger would charge at 500mA?
A quick glance at range of the smaller 'Li-Po' pouches, apparently similar (albeit a different company) to those that you refer to, appear to have a maximum charging current specification that is proportional to their capacity. Looking at one range sold on Amazon (UK) named EEMB shows:
"EEMB Lithium Polymer battery 3.7V 480mAh 502245 Lipo Rechargeable Battery Pack with wire JST Connector" which specifies a maximum charging current of 480 mA.
The full range includes cells with smaller and larger capacities, and for the models I looked at, the maximum charging rate corresponded to 1C ...
My interpretation of the TTGO T-display schematic and TP4054 data sheet suggested that the charging rate was set to 500 mA, which is slightly in excess of the specified maximum for this cell.
Other cells in the same range include a 320 mAh cell, with a corresponding charge rate of 320 mA. Charging this at 500 mA is more than 50% above the cell specification.
I accept many people, like you, will purchase a 500 mAh or larger cell. But, I can foresee someone looking for a minimal size and price, running a low current application, or an application which spends most of its time connected to an external supply and only requires the battery for short periods, who might choose a smaller cell, if they are not provided with clear guidance.
Perhaps, as you suggest, the number of people at risk is very small, but I would like to see good engineering principles being recommended in the hobby world, whenever practical. That includes making situations like this one much clearer.
Best wishes, Dave
@davee NO, I just picked one at random. ANY LiPo 3.7V will work just fine. I would introduce you to the people who understand in excruciating detail how these lithium battery circuits work in UPS mode and normal mode but I don't want to lose them as friends.
Is it best to match the LiPo cell to the charge parameters, obviously yes, but if I value more capacity over on-board charging, then I will buy the higher capacity charger and charge off the board. I am not even convinced the on board charger would not work, albeit slowly. When I had my 7,200 wH battery bank, I tried a number of charge settings. Off the top of my head I think the recommended for long life was 0.5C but if you wanted faster charging at some small cost in lifetime then 1.0C or even 2.0C was ok. Since the nominal (under 8 yr 100% replacement warrantee) lifespan was in excess of 10 years and I was not likely going to live that long or at least have sold them I didn't care if the 10 was reduced to 9 or 8yrs.
EDIT It looks like I failed to post, will do now in tools.
NOTE: I posted something about the 3 SIZES (length) of 18650 in the Tools section a few days ago. All 18650 users would be well advised to understand that.
NOTE2: Based on personal bad experience, when sliding a 18650 cell into a standard holder, place the positive end in first then slide the negative end in. If you do it backwards, you will find some cells have a negative unguarded outer shell that can come in contact with the contact while the positive post is in contact creating a mm length short circuit over a metal strip that makes a very loud noise.
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
Hi Ron @zander,
Thanks for posting the 18650 data ... it fits my somewhat less precise impression, and the other data.
As for the TTGO charger versus cell choice issue, and your assertion "ANY LiPo 3.7V will work just fine. "
I don't think you have realised my concern - it probably got lost in all the pages I type, so let's do it slowly:
Consider someone else buying the TTGO display and a small Li-Po 'at random', much as you did.
But imagine their 'at random' Li-Po was a little smaller in capacity than yours, say the 350 mAH capacity, instead of your 500 mAH one.
e.g. "EEMB 3.7V Lipo Battery 350mAh LP243454 Lithium Polymer Ion Battery Rechargeable lithium ion polymer battery with JST connector" on UK Amazon
This looks very similar to yours ... just 2/3rd of the capacity ... which, we will assume, in terms of its ability to hold charge is sufficient for this user's task.
---------------
Now this device states a maximum charge rate of 350 mAH ... i.e. 1C, which seems to a common specification for this modest Li-Pos
But, according to my analysis** of the published TTGO T-display unit schematic and TP4054 data sheet, it will charge any connected cell at 500 mA.
This is nearly a 50% higher rate than the cell is specified to be able to accept!!
I fear this cell might catastrophically overheat during charging at this rate, assuming the advertised specification is accurate.
Best wishes and take care, Dave
--------------------------------------------
** Footnote **
Analysis based on data sheet from https://media.digikey.com/pdf/Data%20Sheets/UTD%20Semi%20PDFs/TP4054.pdf
which includes current table for different values of RPROG, including 500mA for 2kOhm:
and schematic on GITHUB at https://github.com/Xinyuan-LilyGO/TTGO-T-Display/tree/master/schematic
which shows RPROG value of 2kOhm:
@davee The normal charge rate for this chemistry is 1C, the fast charge rate is 2C. That means any battery from 250mAh up is safe. HOWEVER I was able to find 220mAh LiPo bag cells on Amazon which would be 2.27C. While that is above the 2C rating, it is only 14% higher. Remember that the battery will reach the 4.2V cutoff that much faster, so the 14% greater heat is there for a shorter time. The most dangerous time for a LiPo is trying to charge it when the battery is discharged below 2.5V. Although you could install a technically too small cell, it is not as dangerous as other scenarios like charging below 2.5V
Remember the normal voltage range is 4.2V fully charged, 3.0V is the minimum safe charge with the nominal being 3.7V. NOTE the use of the word safe in minimum, the real danger comes from almost used up cells left on a charger for an extended time often unattended. That 350mAh cell during it's normal safe life will never overheat as much as the now 2.5V 1000mAh cell.
If the point you are so desperate to make is that the consumer has to be aware of the risks, then take your celebration lap. This is one of many daily decisions we make that could cause a disaster. You can't engineer out stupid.
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
@davee I keep forgetting to ask, do your LiPo bag cells contain the protection circuits? All of mine do so the situation you were worried about earlier can't happen. I forget the complete list of protections but as far as this old man's memory goes it's something like
1. Over charge
2. Over discharge
3. Temperature
4. Short circuit
maybe more.
Here is a picture of the IC's also known as PCM (Protection Control Module)
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
Hi Ron @zander,
Thanks for posting the 18650 data ... it fits my somewhat less precise impression, and the other data.
As for the TTGO charger versus cell choice issue, and your assertion "ANY LiPo 3.7V will work just fine. "
I don't think you have realised my concern - it probably got lost in all the pages I type, so let's do it slowly:
Consider someone else buying the TTGO display and a small Li-Po 'at random', much as you did.
But imagine their 'at random' Li-Po was a little smaller in capacity than yours, say the 350 mAH capacity, instead of your 500 mAH one.
e.g. "EEMB 3.7V Lipo Battery 350mAh LP243454 Lithium Polymer Ion Battery Rechargeable lithium ion polymer battery with JST connector" on UK Amazon
This looks very similar to yours ... just 2/3rd of the capacity ... which, we will assume, in terms of its ability to hold charge is sufficient for this user's task.
---------------
Now this device states a maximum charge rate of 350 mAH ... i.e. 1C, which seems to a common specification for this modest Li-Pos
But, according to my analysis** of the published TTGO T-display unit schematic and TP4054 data sheet, it will charge any connected cell at 500 mA.
This is nearly a 50% higher rate than the cell is specified to be able to accept!!
I fear this cell might catastrophically overheat during charging at this rate, assuming the advertised specification is accurate.
Best wishes and take care, Dave
--------------------------------------------
** Footnote **
Analysis based on data sheet from https://media.digikey.com/pdf/Data%20Sheets/UTD%20Semi%20PDFs/TP4054.pdf
which includes current table for different values of RPROG, including 500mA for 2kOhm:
and schematic on GITHUB at https://github.com/Xinyuan-LilyGO/TTGO-T-Display/tree/master/schematic
which shows RPROG value of 2kOhm:
You left out some important information
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
Hi Ron @zander,
I do not have any of this type of cell.
It is possible that the protection circuit with this particular example would prevent a problem, I don't know.
Similarly, I don't know if every manufacturer fits such a control system.
Having secondary protection systems, which do not affect 'healthy' operation, but cover the rare case of breakdown of the main system is usually beneficial ... for example, adding a fuse to protect for a major insulation failure.
But using the protection system as a second level of control for normal usage, because the primary control system is inappropriate, and hence incapable of fulfilling the task, is (in my opinion) very poor practice, because the two control systems will both be actively trying to take control, even though they have not been designed to function in a co-ordinated configuration. (Note, the TP4054 has a relatively complicated control law, as it tries to match the applied voltage and charging rate to the state of charge, analysing the result of having two active mechanisms, is probably non-trivial.)
So, whilst I happily concede that it may work out fine, I still think the TTGO product should be more specific as to the cell requirements, with the appropriate data obviously displayed in the sales adverts.
So, sorry if I have accidentally caused you any distress .... that was, and is never, my intention.
Best wishes, and take care, Dave
@davee Dave, I trust the highly paid engineers who created the designs, and laws that control batteries. If you think you can do better, then by all means present your proposal to the applicable powers that be. In Canada it would involve at least CSA (Canadian Standards Association)
CSA standards are mandatory for electrical products in all Canadian provinces and territories.
and in the USA UL (Underwriters Lab)
UL develops and publishes standards for product safety, security, and sustainability. These standards cover a wide range of topics, including electrical engineering, fire prevention, and chemical injuries.
I am sure Europe has a similar organization.
Be sure to put on your belt AND suspenders when reading this, I wouldn't want you to have your pants fall down tripping you so that you hit your head and ............ damage some furniture. LOL 🤣
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
Hi Ron @zander,
I don't know anything about Canadian standards, and standards are not my expertise generally, but my impression of the standards I am more aware of, is that an item that is to be sold, etc., consisting of a combination of individually certified items, must also be reassessed. Furthermore, depending upon the functionality, etc. of the combined item, relative to that of its component items, then the list of applicable requirements to meet the standards may also be different.
In addition, many standards allow the manufacturer to self-assess their products. I appreciate many manufacturers are diligent and honest, but sceptical that corners are likely to be cut by others.
So, I stand by my previous comments, but leave you the option of being more optimistic than me.
On the positive side, perhaps by chance, the cell you chose is (just) on the acceptable size, so providing you don't start to use a down-sized/down-spec'd cell, it appears your cell meets the expectations of the TP4054, which we both hope, implies it will be fine!
=============================================
I think we have played this topic long enough.
So, I propose we retire to respective corners for a short break, and find a new topic.
Best wishes and take care, Dave
@davee No problem as long as finish with a short reply, everything you just said is either WRONG or just your ..... opinion.!
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.
@davee Somehow you missed a key point re
On the positive side, perhaps by chance, the cell you chose is (just) on the acceptable size, so providing you don't start to use a down-sized/down-spec'd cell, it appears your cell meets the expectations of the TP4054, which we both hope, implies it will be fine!
It doesn't matter, these cells have the protection circuits not the board it is powering.
I just looked at the datasheet for the TP4054 and see that the charger IC automatically terminates the charge (my words when the battery is full) so I can not see any way for the TP4056 to overcharge since both the TP4056 and the cell protection IC will prevent it.
AGAIN, the most common causes of LiPo failure other than physical damage (hammering nails into the cell) are end users not following instructions such as exposing the cell to enough heat to cause runaway, and not using a proper LiPo charger.
First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, 360, fairly knowledge in PC plus numerous MPU's & MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
My personal scorecard is now 1 PC hardware fix (circa 1982), 1 open source fix (at age 82), and 2 zero day bugs in a major OS.