TIDA-060008: Issue on multivibrator

...

Hi Robin,

Do you have a scope shot of the "Q" output during a state change from RX to TX as well as the "D" line signal?  As it does seem like that is the most likely culprit and to help diagnose the issue understanding the input and outputs of the multivibrator will help determine.

The devices that you have chose to replace the original suggested devices are different - the biggest difference existing between the HVD and the THVD devices are speed - as the HVD part is much faster than necessary - but I don't think that should cause the problem you are seeing - realistically the biggest issue would be more higher energy content on the bus - but that shouldn't affect the console side pins.

Please let me know if you can show the "D"  and /A lines and the "Q"/Enable lines between the RS-232 and RS-485 device. 

Best,

Parker Dodson

Hello,
Here is the route that my signal takes:
I send, from a terminal, the number "0".
The first picture shows the signal on the RS485 side (between A and B), then the picture shows the signal at pin "R", then the level of pin "D" and /A. Finally, the last photo shows the pin "Q" at the moment when it activates DE and deactivates RE. Needless to say, the signal at the receiver is no longer "0". When I see this, I feel like my vibrator is affecting my data bus.

Best,

Robin Minaud

Hi Robin,

Is it possible to disconnect the load from A and B and measure the open circuit voltage across during the waveform - I don't think the multivibrator would be causing the signal seen on the A/B signal shown in your last reply.

From what it looks like based on the reference design - when 'D" goes low the multivibrator will output a high pulse switching the RS-485 device into drive mode. This causes "R" to go from a high output to a High-Z output - which may explain why "R" goes low in the scope shot above. If you want the "R" pin held high when it goes high Z (as when the device is in RX mode it will output high during bus idling, open, and short conditions) - This can be done by adding a pull-up to VCC on "R" (4.7K should be fine as the value) 

Essentially I think the "D", "R", and "Q" pins look mostly as I'd expect based on  your design - but the A/B pins do have a bit of variation and I am wondering if the loading on these pins may be somewhat to blame. 

Please let me know!

Best,

Parker Dodson 

Hi Parker,

I disconnected A and B and I left the 120 Ohms resistor between the two. I sent the number "0" on the RS232 side and here is the screen I could observe (see picture 1.A and B  2.R  3.Q). The transmission seems to be good (the signal between A and B seems to me rather scrambled) but I can't test the reception since I don't have the bounce. I also tried sending from the RS485 side and removing the bounce. Again, the transmission is good but I can't check the reception. When I add this bounce to check the transmission and reception, it's then that nothing works anymore. I had also tried with a PULLUP on the 'R' pin and nothing worked, my signal was constantly high.

Best, Robin Minaud

I made a mistake, the third picture is the 'D' pine. See photo below for the 'R' pin. This picture was taken in the configuration where I send from the RS232 side and without bounce. The 'R' pin should not even move in this case. This is not what the scope shows though.

Hi Robin,

The transmission on A and B look much better when removing the rest of the system - I am not sure but it seems like there could be a missing bit however as it does seem the "D" pin sends more bits - I am not sure if that is because Q rise time + driver enable time causes a missed bit.

That being said - The "R" pin looks like the bus it is connected to is discharging when "R" goes high-Z - which explains the slow fall time. But I want to make sure I am understanding - when you have a pull-up on this pin the "R" pin still goes low when the driver is enabled? If yes has this behavior been seen on multiple units? If not can you test another unit - just one or two other ones if possible to see if there could be a potential part issue that may be causing this.

Please let me know!

Best,

Parker Dodson

Hi Parker,

When I put a PULL UP on pin 'R', the pin is always high (no matter if 'R' or 'D' is activated) with very slight variation of 0.1V when sending or receiving. The 'RE' pin is at low level and goes to high level when it is sent, so it is not the pin that puts the 'R' pin in this state. The same goes for the ICL3232 component which cannot influence 'R'. No matter the component and no matter the content sent, I tried to send on the RS485 side and then RS232 but also with another component ICL3232 and SN65HVD485 and it does not change anything, with a PULL UP on the pin 'R', this pin always remains at the high level.

I can't figure out what is causing me this problem.

Best,

Robin Minaud

Hi Robin,

Have you seen this issue on multiple units? 

"R" should go low when the voltage from A to B is < -200mV (minimum) and /RE is low - when the A to B voltage is at 0V "R" will be high if /RE is low. The pin should go high z when /RE goes high - keeping a pull-up will keep the "R" pin from discharging. You couild bias the bus to a -200mV idle voltage between A and B to put "R" in the low state.

The scope shots itself doesn't seem problematic as the "R" pin seems to go low when it is in high-Z. 

Please let me know if you this issue has been seen on multiple units - because right now I am not sure there is any problem based on teh scope shots.

Best,

Parker Dodson