Why is this strange circuit used for data com (multifunction microphone, ham radio)

[Eugen_E]

Hello, I need help with a circuit I saw recently.
In radioamateur transceivers, like ICOM or Xiegu, etc. a "multifunction microphone" with keypad is used for push to talk (PTT), voice and remote commands to be sent to the transceiver. This multifunction microphone has to have proper EMI performance, so powerful RF signals don't affect it, and also key presses won't influence the sensitive receiver... so capacitor and inductor filters are used at both ends and also transient suppressors.

This microphone is connected with a 8 wire cable ~ 1 meter in length, similar to Ethernet cable, with modular connectors. Power and ground, audio from microphone on 2 wires to avoid ground loops, PTT, etc. signals are sent, and also a data signal for commands generated when pressing keys.

Here is a portion of the "multifunction microphone" circuit schematic,

and the transceiver circuit it connects to (from a Xiegu G90 transceiver):

MIC_DATA is a data signal (serial) generated by a microcontroller inside the "multifunction microphone", and sent through that strange circuit in red.
In the tranceiver, a comparator (in red) is used to recover the MDATA signal.

Questions:
- Why is that circuit with high value resistors used, instead of driving it directly with the microcontroler pin or some driver circuit designed for the cable impedance?
- The circuit for data transmission is sensitive to the capacitance of the cable used? And perfectly only with the OEM cable?

It works perfectly only during RX with CAT-5, CAT-5E, CAT-6 cable, if the pairs are crimped according to TIA/EIA568A and TIA/EIA568B.
But, when transmitting, RF is picked up by the MIC signals because the MIC signal wires are in different pairs... so a CAT7 or better, shielded cable is needed.

To avoid RF pickup, I tried to put the MIC wires in the same pair... but to my surprise, the data communication is not working anymore, maybe because it is expecting a specific cable capacitance...

Thanks

 

[barry]

- Why is that circuit with high value resistors used, instead of driving it directly with the microcontroler pin or some driver circuit designed for the cable impedance?

Why do you think those resistors are “high value”? Driving directly from the microcontroller is a bad idea. It looks like there actually IS “some driver circuit”; that’s what those transistors are.

It’s not clear from your post, but are you replacing the original cable, causing it to not work? Not sure what you mean by “signal wires are in different pairs”. it looks like these are all single-ended signals.

 

[Eugen_E]

It’s not clear from your post, but are you replacing the original cable, causing it to not work?

Yes, I was trying to replace the original cable which is very hard and springy with my own cable. It worked with standard cat 5 Ethernet cable but it picked RF on TX. So I tried to put everything in pairs 1-2, 3-4, 5-6 is the microphone. To my surprise, the data signal is not received now.

People had success with good quality cat 7 or 8 cable, shielded and wired like standard Ethernet.

Regarding the driver circuit, i think the microcontroler could drive through 8k2 and 1.5 k resistors without those digital transistors... Anyway, that circuit with many resistors and a voltage divider is quite strange

 

[KlausST]

tried to put everything in pairs 1-2, 3-4, 5-6

You showed a CAT 5 connector.
Here the pairs are not as you showed.
They are: (1 -2), (3-6), (5-4), (7-8)

But I´m not sure whether the use of CAT.5 and/or paired cables are good at all.

Klaus

 

[Eugen_E]

You showed a CAT 5 connector.
Here the pairs are not as you showed.
They are: (1 -2), (3-6), (5-4), (7-8)

But I´m not sure whether the use of CAT.5 and/or paired cables are good at all.

Klaus

Yes, initialy it was 1-2, 3-6, 5-4, 7-8, but it picked RF on MIC.
So then i tried to crimp it 1-2, 3-4, 5-6, 7-8, and data transmission stoped working.

 

[KlausST]

Hi,

until now you gave nothing we can validate. The schematics show how it should be, the text how you want it to be.

If you want us to validate your solution, then you need to show us YOUR artwork, like photos of your modifications, additionally maybe YOUR wiring diagram.. and so on.

****
V / The strage COM circuit really is a bit strange. They modfied the 0V/5V levels to 1.75V/2.4V levels and did a bit of slew rate limitation. I guess their idea was to reduce transmitted noise levels.

But relying on two seperate power supply levels ... on different PCBs is not very reliable.
I guess a true RS458 or LVDS would be less noisy, more reliable, lower part count ....

Klaus

 

[Eugen_E]

I don't have a solution, i just tested cables and was puzzled by the results. Did they design the circuit to work only with the original cable?
I was expecting maybe data errors during TX...