TM1637

[dim78]

has anyone work on TM1637 display with a 18f4550 or a other microchip model ?? a simple code for it with oshonsoft language??

 

[For The Popcorn]

One thing to watch out for – if you're using some of the modules with display and TM1637 chip :

At least some of the modules have capacitors on the clock, DIO and STB lines that are too big, killing the square wave pulses. If you're having trouble getting anything to work, remove those three caps. They are not normally needed anyway.

 

[DogFlu66]

I didn't know about those modules, so I ordered a couple of them and downloaded the Arduino library to see how they work.

 

[Nigel Goodwin]

They work fine, I've converted code for PIC's to use them.

 

[For The Popcorn]

Here is my tale of woe regarding TM1637 modules. Ignore my above comments at your own peril, even if Nigel Goodwin is certain "that they work fine." In at least some cases too large of filter caps have been installed. This was the case with the modules I purchased, and similar results have been reported in Arduino forums where the modules only operated at extremely reduced clock rates.

 

[For The Popcorn]

Here is the pertinent message in the above long thread documenting the capacitor issue in at least some of the TM1637 modules. It may not be the problem in this case, but if you think everything in your code is correct and it still doesn't work, this is something to check.

There's also code for Swordfish Basic in this thread which may be helpful.

 

[Nigel Goodwin]

Here is my tale of woe regarding TM1637 modules. Ignore my above comments at your own peril, even if Nigel Goodwin is certain "that they work fine." In at least some cases too large of filter caps have been installed. This was the case with the modules I purchased, and similar results have been reported in Arduino forums where the modules only operated at extremely reduced clock rates.

I too read the tales of woe - and tried them both with, and without, the capacitors - in both cases they worked fine (at least while using a PIC).

I certainly agree the capacitors seemed much too high, if you wanted to add capacitors across them at all?.

Try them with, if it doesn't work, them remove them - if it still doesn't work (as is likely), then dig deeper - if it does work, then smile, job done!

 

[For The Popcorn]

I too read the tales of woe - and tried them both with, and without, the capacitors - in both cases they worked fine (at least while using a PIC).

So the modules you had worked fine with or without the caps, and by that you make the blanket statement "They work fine", discounting any possibility that others' experiences are true?

No. What you proved by your test was the caps on the boards you have were of the proper value.

I'm not trying to be argumentative, but your tendency to discount any experience that doesn't agree with your own isn't entirely helpful. (smiley face makes this non-rude).

 

[Nigel Goodwin]

So the modules you had worked fine with or without the caps, and by that you make the blanket statement "They work fine", discounting any possibility that others' experiences are true?

No. What you proved by your test was the caps on the boards you have were of the proper value.

I'm not trying to be argumentative, but your tendency to discount any experience that doesn't agree with your own isn't entirely helpful. (smiley face makes this non-rude).

Hardly, I clearly said "try it with the caps, and remove them if it doesn't work" - I tried three different types of boards, all worked, but on one I removed the capacitors (and measured them) and confirmed it worked fine without them. On that board at least, the capacitors were the 'wrong' high value - I didn't bother trying the other boards.

 

[tumbleweed]

As long as you use clock timing 100x greater than the TM1637 datasheet shows the chinese modules I had "work fine".
Most of the arduino code I found uses 100us, while the chip itself works at < 1us once you remove the offending caps.

 

[For The Popcorn]

Thanks again for the help with that tumbleweed !

 

[Tony Stewart]

Looking at datasheet specs
Rout = 100 Ohms approx @ 5V based on Vol/Io
CL= 100 pF times 100 ohms = 10 ns for fall time

But with passive pullup risetime 10k * 100pF = 10 us risetime on a period , the 100 pF will be OK only if external driver is 50 to 100 ohms and not I2C open drain/collector. Then C must be reduced to < 1us if EMI is a problem. Considering twisted pair is about <=100pF/m this will limit speed and provides the maximum capacitance suggested.

So the driver impedance, clock/data rate & cable length determine maximum C permitted.

 

[For The Popcorn]

Looking at datasheet specs
Rout = 100 Ohms approx @ 5V based on Vol/Io
CL= 100 pF times 100 ohms = 10 ns so I expect no problems for < = 10 Mb/s perhaps more

View attachment 144654

Probably true, IF THE CORRECT CAPACITORS ARE INSTALLED. This was NOT THE CASE in the modules I and others have had.

I posted a warning of a known problem. Not always the problem, but a possible problem to check for.

Now, both you and Nigel have stated this isn't a problem. Well done.

 

[Nigel Goodwin]

Probably true, IF THE CORRECT CAPACITORS ARE INSTALLED. This was NOT THE CASE in the modules I and others have had.

I posted a warning of a known problem. Not always the problem, but a possible problem to check for.

Now, both you and Nigel have stated this isn't a problem. Well done.

As I've clearly repeated, if it doesn't work, try removing the caps - no need to remove them if it's going to work anyway.

 

[Tony Stewart]

Probably true, IF THE CORRECT CAPACITORS ARE INSTALLED. This was NOT THE CASE in the modules I and others have had.

I posted a warning of a known problem. Not always the problem, but a possible problem to check for.

Now, both you and Nigel have stated this isn't a problem. Well done.

Please do not misunderstand what I wrote.
If using I2C even a 1 m cable is a problem at 500 kHz with no capacitor and a 100 pF twisted pair cable with 10k pullup only. But using a 100 or 50 ohm driver is no problem.

 

[DogFlu66]

Hello; I received some TM1637 modules and already started them on oschonsoft, they are two different versions, one works with 1uS delay. and others with 30uS. Of course, I had to write the library for oschonsoft, based on C libraries and TM1637 datasheets.

Does anyone know which module I should order with a keyboard, so that they are compatible with the ones I have without a keyboard?. Well the ones I find are TM1638 and those seem to me to use a real I2C, or am I wrong?.

 

[For The Popcorn]

As I stated previously, check the capacitors installed between the clock and data lines. Some modules have too large of value.

Either change them to the value on the datasheet or REMOVE them. They are usually not needed.

 

[Tony Stewart]

The TM1637 is not I2C and data must not change when CLK is high. This means data out should change after falling edge of CLK and then is saved in TM1637 whenever CLK is high. So it is not edge triggered.

" When data is input,DIO signal should not change for high level CLK and DIO signal should change for low level CLK signal"

 

[DogFlu66]

In reality, they are not modules that I plan to use in any work project, it is just the challenge of getting them up and running. In any case, I will do what you indicate and remove the capacitors from some of them to see what happens.

 

[For The Popcorn]

Only the 2 caps on the clock and data.....