Notifications
Clear all

TTGO T-Display Issues

27 Posts
3 Users
1 Reactions
343 Views
(@ethanwelds24)
Member
Joined: 2 months ago
Posts: 3
Topic starter  

I am currently working on the Mecanum Car Project. 

I have the car and remote working properly with no apparent issues. I am using the TTGO T-display for the controller and was able to get its MAC address and successfully loaded the code for the controller and control the car reasonably well. 

My issue is that I am now unable to get the T-Display to be recognized by my computer. I use the same USB-C to USB-C cable that I have been using for it up to this point and there is no activity on the board (no blue light and no sketch running). When I hook up that same T-Display to a battery it turns on and runs the car remote sketch without any issues. It seems as if I am stuck with a permanent TTGO T-Display remote. I have wateched the video about flashing a "bricked" T-Display but I cannot even get mine to be recognized in a USB port.

I'm confident that someone else has run into this issue and that it is a simple fix. 

Statement of Facts:

-The cable is a properly functioning data cable and I have tried a few on various device to ensure that they work.

-The computer is running Windows 10

-The board has any extraneous wires disconnected from it when I attempt to get my computer to recognize it.

-The Computer - Cable - T-display combo has successfully connected and uploaded a number of different sketches up until I loaded the mecanum remote sketch.

-When connected to a battery the display powers up and runs the mecanum remote sketch


   
Quote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 4 years ago
Posts: 8046
 

@ethanwelds24 Windows device management mess ups is why I switched to Mac a few years back. Try powering off the windows box, leaving the USB cable disconnected. Then try again.

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.


   
ReplyQuote
(@davee)
Member
Joined: 4 years ago
Posts: 1924
 

Hi @ethanwelds24,

  I don't have a TTGO T-display, but my first approach would be to ensure the TTGO is getting still getting power from the PC, via the USB cable.

From the diagram at https://lilygo.cc/products/lilygo%C2%AE-ttgo-t-display-1-14-inch-lcd-esp32-control-board

image

The GND (0V), 5V and 3.3V are shown on the bottom two rows of pins.

The USB cable should be plugged in, but no other source of power should be connected to the TTGO. The PC should be booted. Use a multimeter to measure voltage from (A) 5V pin to GND pin , and (B) 3.3V pin to GND pin.

I would expect to see 5V for test (A) and 3.3V for test (B), regardless of the software installed on the TTGO or the drivers loaded into Windows.

Then please report your findings. Hopefully someone will be able to offer some advice to proceed further, depending upon these results.

Best wishes and good luck, Dave


   
Ron reacted
ReplyQuote
(@ethanwelds24)
Member
Joined: 2 months ago
Posts: 3
Topic starter  

@zander Interestingly enough I started all of this on my Mac, downloaded the driver, made a couple fun sketches for my ttgo (it went well) but I upgraded to Sequoia about the same time I did the mecanum remote sketch and now I get nothing out of the ttgo


   
ReplyQuote
(@ethanwelds24)
Member
Joined: 2 months ago
Posts: 3
Topic starter  

@davee Did some testing, I get no voltage out of the pins (3v to G or 5V to G) but I do have voltage coming into the board when I use my tiny little probe to test the small solder spots coming out of the USB connector on the board.


   
ReplyQuote
(@davee)
Member
Joined: 4 years ago
Posts: 1924
 

Hi @ethanwelds24,

  That is unusual, but consistent with the board playing dead.

  The TTGO reference I gave in the previous message suggests this GITHUB address for more tech support:

https://github.com/Xinyuan-LilyGO/TTGO-T-Display

On which I found a schematic for the board:

https://github.com/Xinyuan-LilyGO/TTGO-T-Display/blob/master/schematic/ESP32-TFT(6-26).pdf

From this, I offer a few snips to show the incoming USB power path:

1/ The USB-C connector seems to be labelled Typey-C ... and I presume you managed to find

5V on the pins labelled PWR PWR1 UBUS1 and UBUS2, all of which go to net VBUS

with respect to the 4 GND pins that go to the GND net.

(BTW Click on this image, and the following ones, to make them display at a readable size.)

image

2/ GND net is probably a ground plane, which is usually robust,

so following the VBUS net ...

image

this passes through a diode D1 IN4148 type, and on to a switching circuit, but that part of the circuit will not provide power from the USB to the main board, so need to find another route using VBUS net.

3/ This snip shows a diode D2 connecting VBUS to +5V net:

image

Assuming VBAT is not connected to a supply, then D2 should conduct and a voltage of just under 5V ... say about 4.7V ... should appear at the output on net +5V.

4/ At this point, return to the snip used for 2/ (copied here again for convenience), noting +5V net is on lower left side, where it feeds a 5V to 3.3V voltage regulator:

image

and +5V also appears on the 0.1" pitch connectors that you have noted, do not exhibit any power:

image

At this point, I think all I can do is to leave you to trace the same path through, to determine where the volts disappear.

As you say +5V is missing, then obviously the 3.3V regulator will be unpowered, so I would start by concentrating on the VBUS and +5V path.

D2 is a prime suspect, but as in the best detective stories, the prime suspect is not always the guilty party, so keep an open mind.

-----------

If you find a faulty part, then you might wonder if it just failed because it wanted to, or did you accidentally 'push it'? 

Unfortunately, this is often a difficult question to answer.

I just note that many of the ESP32 development boards, this one included, has some switching components to try to protect the circuit, and also the development PC, if power is applied via the USB and from another source, at the same time. I have not analysed this particular circuit, but in the past, I have noted the protection system did not look very robust.

My first concern has always been that the PC is vulnerable to damage if a voltage appreciably greater than 5V is accidentally applied to its USB 5V power line ... to the extent that the machine may be unrepairable. Of course, loss of a development board is also very undesirable, although usually, not quite so painful to the wallet.

Please do not think, I am in any way insinuating, that this has occurred in your case ... I am merely pointing out the need to be careful, especially if the project has any power sources, including batteries, that provide more than 5V.

----------

Good luck, please ask further questions if necessary, and if you find the culprit, then please post that information.

Best wishes, Dave 


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 4 years ago
Posts: 8046
 

@ethanwelds24 That sounds like a dead duck. Any chance there is a fuse @davee I know some boards have them.

EDIT I just noticed something, is that a USB-C connector? If it is, try turning it upside down, if that fails, try another USB-C cable preferably an expensive high quality type, but just make sure it is working somewhere else. I have a couple testers now for ALL USB connectors, one a simple continuity, the other checks signals. Not cheap but they pay for themselves pretty quick. Let me know if you want info on them, they are not the usual you will find, one is a open source made in Poland.

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.


   
ReplyQuote
(@davee)
Member
Joined: 4 years ago
Posts: 1924
 

Hi Ron @zander,

 I haven't seen any ESP32xxxx boards with fuses, and I didn't spot one on the schematic for that one.

A USB cable or connector fault seemed unlikely, since @ethanwelds24 reported "Did some testing, I get no voltage out of the pins (3v to G or 5V to G) but I do have voltage coming into the board when I use my tiny little probe to test the small solder spots coming out of the USB connector on the board."

But this is an on-going investigation, and experience tells me that it is easy (for me at least) to be confused by a single observation, so I always try to double-check any measurements, not just by repeating the first measurement, but by looking for a second different measurement, before jumping to conclusions.

  I would like to think the board is fixable by replacing one or two minor components. Of course, that can present practical issues, such as being able to solder rather small parts, which not everyone will be equipped to do. I hope my reply is helpful, if only to establish what has failed, and possibly to suggest why it probably failed, in case there is cause, that can be addressed.

Best wishes, Dave


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 4 years ago
Posts: 8046
 

@davee Just checking Dave, some boards do have tiny fuses. As far as @ethanwelds24 testing for the presence of 5V NEAR the USB pad, I know from past experience there are often multiple solder pads and traces with power flowing every which way. 

-When connected to a battery the display powers up and runs the mecanum remote sketch

Where is the battery connected? I assume there is a VIN pin in order for it to work at all. If that is how it is being connected, then the list of possible damaged components is fairly small. Whether Ethan can repair it is a separate matter. 

OR

Is this a board with a grove connector on the back for a LiPO battery? If that is the case, it should be possible to measure the charging voltage if any at the grove connector. Just one more clue.

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.


   
ReplyQuote
(@davee)
Member
Joined: 4 years ago
Posts: 1924
 

Hi Ron @zander,

  Some boards using USB for power input or output do indeed have 'resettable fuses', namely positive temperature coefficient thermistors, whose resistance rises if the circuit attempts to pass an excessive current, but I do not recall seeing one on any of the ESP32 development board schematics, including, not being present on this one.

   I can't see a battery holder, in the photos of the TTGO advert I referenced, just a 2-pin connector and pair of wires, to mate with the socket on the board. The schematic I referenced has a 2 pin connector, which goes to TP4054, which is a linear charger chip for a single lithium-ion cell.

It appears that the TP4054 is able to use the USB 5V line, as a source of power to charge the cell.

If only a single cell is attached to the 2-pin connector, then a single cell voltage should not present a hazard to the PC. My concern about damaging the PC arises for projects with battery or other supply voltages higher than 5V, which may accidentally become connected to the USB 'VBUS', if only for a moment, due to the slip of a test probe, etc. Motorised projects, in particular, often need more than a single cell as a power source.

I haven't carefully examined the logic built in to the TP4054 cell charger chip. I am not sure if it will output a voltage when the cell is not connected. Hence, looking for a voltage on the cell connector might give a misleading result. Perhaps probing the power input pin to the charger chip would avoid this possible ambiguity.

I presented the information above, in the hope that a 'simple' repair might be possible.

Best wishes, Dave


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 4 years ago
Posts: 8046
 

@davee I did NOT say battery holder, in fact I have never seen a holder for bag type LiPO batteries. I used your link and here is the grove connector. I strongly suggest to @ethanwelds24 that he meter those pins when the USB is plugged in. The other reason I knew it must exist is on at least one of the diagrams you posted it said VBAT, it also had a voltage ref that was exactly the LiPO battery voltage. Pretty big clue Dave, I was worried about you before after some odd posts, now I am really worried.

If a bad component can be found, @ethanwelds24 will likely need a microscope, and possibly those hot tweezers or at the very least a heat gun with variable temp and air flow. Maybe we should wait to find out if he wants to spend the money on those as they are both expensive vs just buying another board.

Screenshot 2024 12 25 at 12.30.49
Screenshot 2024 12 25 at 12.35.25
Screenshot 2024 12 25 at 12.36.08

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.


   
ReplyQuote
(@davee)
Member
Joined: 4 years ago
Posts: 1924
 

Hi Ron @zander,

  Sorry about the mention of battery holder ... although I have never used one, I have seen adverts for other development boards with an 18650 cell holder, including charger arrangement, etc., so that is what I imagined you were referring to.

  Clearly, the inclusion of the particular charger chip shown on the schematic, implies the board is meant to support a lithium-ion single cell, as a power source.

Whilst I often see similar voltage charging curves regarding lithium-ion and lithium-polymer cells, I am aware that a number of different cathode types of lithium-ion cells are in production, and some have different requirements. I have not read all the literature about this particular TTGO board, but so far, I have not seen any specific advice regarding the type of cell to be connected to the bare wire ends. I briefly looked at the data sheet for the chip type I found on the schematic, and whilst it details the voltage and current behaviour of the chip, it is left to the designer to find a suitable cell. I suspect suitable cylindrical cells and pouch-shaped cells are available, implying either format could be used, but given the inherent dangers of misusing lithium-ion cells, I am not making any recommendations of suitable cells.

-------------

I don't know what odd post you are referring to...

My post on attempting to follow the 5V path from the USB to the ESP32 was only an analysis of the path the current had to take, in the hope it could be probed, and the fault located. Until the fault is known, the difficulty of actually repairing the board is unknown, and obviously @ethanwelds24 will have to decide the best course of action, depending on availability of equipment, skills and so on. As I mentioned before, when a fault occurs, sometimes it is valuable to discover the cause. Of course, this is not always possible, and sometimes a component fails due to a manufacturing defect or other random event.

As for equipment likely to be needed, then a small hot air gun is often useful, although a few people are skilled enough to solder even very small parts with a 'conventional' style soldering iron. I also use a cheap magnifying glass, but haven't yet had to buy a microscope. I wouldn't have bought either merely to repair the faulty ESP32 board I had ...  I already had them when the faulty board appeared, and I have used them both many times for other jobs as well.

------

In summary, all I suggested above is a route to try to discover the problem, which may or may not be successful.

If a faulty part is located, then the best way of overcoming the fault is a subsequent decision.

Even if a faulty part is discovered, buying a replacement board may still be the best solution. However, I think it is much better to locate a fault, than to simply become a 'board jockey' who simply swaps parts in the hope of eventually finding a combination that works. Others are entitled to alternate views.

Best wishes and good luck to all, Dave


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 4 years ago
Posts: 8046
 

@davee The LiPo battery is favoured by a few 3rd party manufacturers like TTGO, Waveshare, and SEEEDstudio plus others. They are usually more compact than the 18650 and more powerful. LiPO and LiIon need proper handling for sure, I even had an ANKER powerbank suddenly swell up even though it had not been used or charged in months. With 18650 cells you have the option of LiFePO4 which is extremely safe but less power dense.

I have no idea where you got this

I have not seen any specific advice regarding the type of cell to be connected to the bare wire ends.

the board has a JST connector, I think I said grove in error, certainly not bare wires. I have 4 or 5 of these batteries sitting on my workbench right now.

As far as the following,

I suspect suitable cylindrical cells and pouch-shaped cells are available, implying either format could be used,

NO, the JST connector is for pouch style LiPo cells, the cylindrical 18650 cells are much bigger than the board itself which makes zero sense and they need a cradle with contacts at each end. 

For clarity here is a link to a LiPo battery showing the JST connector https://amz.cx/3GEv and here is a picture of a board that takes advantage of the 18650 size. Much bulkier package.

Screenshot 2024 12 25 at 15.16.16

 

 

 

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.


   
ReplyQuote
(@davee)
Member
Joined: 4 years ago
Posts: 1924
 

Hi Ron @zander,

I agree with most of your reply, but I would like to clarify something.

So far as I am aware, many of the commonly available Li-Pos and 18650s are basically the same underlying chemistry, (albeit many do not have their type clearly printed on the package), and have similar voltage characteristics.

The current characteristics are a combination of physical and construction, so the current characteristics are more varied, especially when pouches are used.

However, as you point out, other chemistry variants including LiFePO4, are also available, and as you know, the operating voltage range varies with the cathode chemistry. e.g. LiFePO4 has a much lower operating range than its cobalt oxide based cousin.

The different voltage variants require different voltage charging characteristics. In addition, the recommended maximum charging rates, etc. vary with both chemistry and physical construction design.

(Of course, some more expensive charger units include a method of setting the charger output characteristics to match the specific cell/battery type, but this capability is much less prominent in the relatively low-cost development boards we often discuss for hobby projects.)

----------

For connections to the cell, the TTGO advert simply shows a simple 2-pin connector to match the 2-pin connector on the board, with short wires terminating in bare ends. The JST connector provides a connection to the board. In principle, the bare wire ends at the other could be attached to any battery or cell.

Perhaps, I missed it, but I couldn't see any mention of the requirements of a suitable cell to attach in either the advert or the accompanying GITHUB page. I do not believe using a 2-pin JST connector on the TTGO board, accurately suggests to the user, what type of cell would be suitable for connection to the free wire ends. I realise some Li-Po cells are sold with this style of connector, but JST connectors are also widely used elsewhere. They are not a 'battery' special connector.

-------------------

Meanwhile, the cell charging chip on the board, is listed as TP4054 on the schematic. The data sheet shows that maximum charging current is set by a resistor on the board, whilst maximum charging voltage is predetermined to be 4.2V by the chip design. I wonder how many people buying this board will read the schematic, and then the TP4054 data sheet (which is quite challenging to understand), and then choose a cell based on that analysis?

--------------------------------

Hence, I happily accept your comments about a Li-Po in pouch format maybe preferred to an 18650 cell on grounds of size and ease of mounting, but I don't think that addresses the electrical matching requirements, which for me at least, is of far greater concern.

I also accept, that some readily available choices may, by chance, be suitable, but others are not.

So I am concerned that this factor is not made much clearer.

Best wishes, Dave


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 4 years ago
Posts: 8046
 

@davee The 18650 has a second chemistry (other than a bunch of special purpose we are only talking LiIon, LiPo, and LiFePO4) which is the very safe but much reduced power density the LiFePO4. These are so safe many RVers store them under their bed. All legit cells have the chemistry printed on the container be it round or bag. LiFePO4 is 3.2V thus 4 make 12.8 so they are sold in car size batteries as they have the same voltage. The others we encounter as hobbyists are 3.6 or 3.7.

As far as pouches and capacity (mAh you will find that there is a variety for some sizes so you can pick the one that fits. There is a shape for esp32 boards for instance and thinner but wider for other use cases.

SIDENOTE, I have camped in COBALT Ontario. The silver mines closed down a long time ago, but they are now re-opening the Cobalt mines due to the world's demand for it.

I only use professional chargers, so I have a SkyRC MC3000 for all my cylindrical batteries of all chemistries MC3000  and an iMAX B6 Mini Charger for the LiPo bags iMAX B6 Mini as well as for any small multi cell batteries that need a balance connection. For the big toys that use 5,000 mAh and up multi cell LiPo like THESE I use a a bigger version of the B6 Mini called the D200neo  It's people who try to charge Lithium batteries without using a properly engineered charger that have the fires.

As far as the following, Dave, get real, the people buying the boards know what kind of battery to get, they look like this LiPo

For connections to the cell, the TTGO advert simply shows a simple 2-pin connector to match the 2-pin connector on the board, with short wires terminating in bare ends. The JST connector provides a connection to the board. In principle, the bare wire ends at the other could be attached to any battery or cell.

Perhaps, I missed it, but I couldn't see any mention of the requirements of a suitable cell to attach in either the advert or the accompanying GITHUB page. I do not believe using a 2-pin JST connector on the TTGO board, accurately suggests to the user, what type of cell would be suitable for connection to the free wire ends. I realise some Li-Po cells are sold with this style of connector, but JST connectors are also widely used elsewhere. They are not a 'battery' special connector.

You are so maddening, the following is hogwash.

Meanwhile, the cell charging chip on the board, is listed as TP4054 on the schematic. The data sheet shows that maximum charging current is set by a resistor on the board, whilst maximum charging voltage is predetermined to be 4.2V by the chip design. I wonder how many people buying this board will read the schematic, and then the TP4054 data sheet (which is quite challenging to understand), and then choose a cell based on that analysis?

It's the equivalent of asking every soccer mom to read datasheets and examine circuitry to make sure she gets the right battery for her flashlight.

THEY ONLY COME IN ONE VOLTAGE (actually multiples of 3.7V - 4.2V) 

DO NOT POST BACK WITH SOME WEIRD VOLTAGE THAT NOBODY USES FOR RC MODELS OR ESP32 OR ARDUINO ETC. 

Screenshot 2024 12 25 at 19.48.02

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.


   
ReplyQuote
Page 1 / 2