Differential TX driver for TDI?

[eduardo1979]

Hi Algan.
I've attached a timing diagram that I think we need. I'm not entirely sure if it will work or not. I'd appreciate the experts' opinion.

[pechkata]

Hi Algan.
I've attached a timing diagram that I think we need. I'm not entirely sure if it will work or not. I'd appreciate the experts' opinion.

Hello Eduardo

At this frequency of 100 microseconds, pumping cannot work.
Or will it consume 400/500 milliamps of current?
You watch Carl's video again.

It's written in this code.

[algan]

Hello, Eduardo, my colleague Pechkata correctly wrote that the consumption will be very high this way, but this version of my first circuit, look carefully at how I drew it, the samples and they overlap with the second pulse, that's why Carl wrote that both should be positive, it is better to make the second circuit there is no overlap of the sample with the pulse, I will see if I can correct the code and I need to reduce the width to 50 microseconds, done this way it will work, and it will also be possible to make one pulse low in the second version or like Carl's bipolar, this variant will even work there too

[boilcoil]

I think you don't need anything more, just use what Karl posted. You already have a schematic, a board and the necessary microprocessor -just do it.
I ported the code to a PIC16F1705 microprocessor, because I have a simulation model for it and at first glance - everything is fine.

[boilcoil]

.........my colleague Pechkata correctly wrote that the consumption will be very high this way........

Once again, please use what Karl wrote. He has anticipated your concerns and at higher frequencies the pulse width is adjusted/reduced.

[algan]

Hi Eduardo, this is what I need too, this will help you do the TDI test. One of the pulses could be negative, both could be positive, there is no way to predict how it will work but, only a real test would give an answer.

[Carl-NC]

This is what the timing should look like:



With bipolar pulsing you no longer need the EFE samples. So now Tgt & Gnd can be called TgtP & GndP, and TgtEfe/GndEfe can be called TgtN/GndN.

[Carl-NC]

Here is modified code for the 16F684 that will compile but I don't know how correct it is. Most likely the Timing_Update() function will need some attention. This is intended to roughly replicate the TDI (not TDI-SL) stock micro; converting it to a bipolar version should be fairly easy and I will try to get to that soon. It runs on the internal 8MHz oscillator so is a bit more limited in timing resolution; using a 20MHz crystal would help.

If you are literally wanting to convert a TDI-SL to bipolar the 16F684 won't do it; it's a 14-pin micro and the SL requires a 28-pin micro.

TDI_Std_16F684.zip

Edit: I started looking at bipolar changes and already found some errors in this code. I'll fix it as soon as I can.

[pechkata]

Hello Carl-NC
Does this diagram not match this code?

[Carl-NC]

Yes.

[algan]

Hi, Carl, the file doesn't compile and gives me these errors.
13 | #include "Voodoo.h"
| ^~~~~~~~~~
1 error generated.
make[2]: *** [build/default/production/_ext/38421408/AppDSP.p1] Error 1
make[2]: *** Waiting for unfinished jobs....
nbproject/Makefile-default.mk:139: recipe for target 'build/default/production/_ext/38421408/AppDSP.p1' failed
make[1]: *** [.build-conf] Error 2
make: *** [.build-impl] Error 2
make[2]: Leaving directory 'C:/Users/algan/MPLABXProjects/TDI_Std_16F684.X'
nbproject/Makefile-default.mk:91: recipe for target '.build-conf' failed
make[1]: Leaving directory 'C:/Users/algan/MPLABXProjects/TDI_Std_16F684.X'
nbproject/Makefile-impl.mk:39: recipe for target '.build-impl' failed

BUILD FAILED (exit value 2, total time: 1s)

[Carl-NC]

Sorry, delete AppDSP.h and AppDSP.c. I did not mean to include those files.

[algan]

Yes, Carl, after I deleted them, it compiled and works, but I want to ask you if there is a possibility to control the frequency with a trimmer. Thanks.

[Carl-NC]

The code reads 2 pots; one for delay and one for frequency. The frequency pot can be a panel control or a trimmer.

[algan]

Thank you very much Carl..I tested the output on pin 13 and the output on pin 3 against 0 and 5 volts, I connected the other two legs of the potentiometer, it doesn't work like that.