• State
  • Replies 3 replies
  • Subscribers 34 subscribers
  • Views 128 views
  • Users 0 members are here
Support feedback
Options
Options
Related

TSS721A: TSS721A with optocoupler ever done successful?

Matthias Lipinsky
Prodigy 10 points

Part Number: TSS721A

Hi.

I dont know if it helps, but I feel the need to talk.

I  am trying to design a M-Bus slave which is galvanically isolated.

Try1:

This was my first try. As soon as I placed/added R12/R13, to digital isolator IC3 went "active" and draws too much supply current. So the 3V3 supply from VDD (Pin11) breaks down.
image.png

So I went back to the datasheet and did another try, closer to the datasheet example (figure10, basic optocoupler design)


Try2:
I searched an optocoupler with very low diode current. I am using LDA111, which seems to work with less than 1mA of photo diode current.
image.png

The parts around OK1 and OK2 seems to work properly. Only issue I had, is, that I need to a pretty small R6 as pull up to create kind of rectangular wave form. But at µC side, current is no issue. So my µC is able to receive data from the M-Bus.

But as soon as I added the µC-transmit via M-Bus path (OK3) I have the same issue. R3 must be very small to create proper waveform. But this draws too much current from VDD (Pin11). So the 3V3iso breaks down again.


My problem now is:
1. digital isolators (try1) needing too much supply current.
2. optocouplers either need too much led diode current and/or too much current thru pull up resistor to recover waveform

And with "too much" I am talking about more than 1.5mA, which can be delivered by VDD pin permanently. (In my opinion, the STC pin and the VDD pin can deliver together 1.5mA)


So the datasheet recommends values for different operation modes (e.g. figure 8). But the optocoupler example does not recommend any value at all.

So serious question:
Does somebody ever achived to design an optocoupler application? How to do this with max.1,5mA total usable current?
What optocoupler and values for R4 and R2 are recommended?
image.png

Thank you
Matthias





  • 0
    Prodigy 10 points

    So nobody able to say something?

  • 0 in reply to Matthias Lipinsky
    TI__Guru 74602 points

    Matthias,

    My apologies for the delay in response. Unfortunately, we don't have any specific reference designs using optocouplers with this device, and since this is an older device, the original development team no longer are a part of TI. Can you please share any waveforms you've captured? The resistor size/increased current shouldn't be necessary to create the correct waveform here. That is, what is recommended should be correct.

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    Prodigy 10 points

    Hi Eric,

    Thanks for responding. As this is an older device, are there any newer version or successors? I wasnt able to find some?

    I didnt saved the screenshots from the waveforms, but it look like the typical capacitor charging curves. I did not investigate it deeper, but I believe, a combination of high CTR, darlington and high output capacitance might be the reason.

    The datasheet from my try states this. 500Ohms at 3V3 is already >6mA. And the 345µs rise time is way too low, even for 2400bd (T=420µs)



    So I am continueing to find an optocoupler/insulator which would work.