Hi
I have a differential signal which will be transmitted from a system (transmits and receives the data through 2 wires only, there is no GND wire).The differential signal has 3 voltage levels -12V , 0V and +12V. I have measured this differential signal with the help of oscilloscope by keeping one wire to GND of oscilloscope and one wire to positive of Oscilloscope.
The differential signal was shown below.

I need to decode this data i.e, convert to single ended signal but here the issue is when this differential is converted to single ended signal using AM26LS31CDBR IC, 0V logic level is either converted to High or low, but it is not following the same logic i.e. High or low.
I need a transceiver which converts this three state differential signal to single ended signal and it also should convert the same converted single ended signal to differential which should be same as three state differential input signal .

Can anyone suggest me the solution for reconstructing the differential signal.

 

What is the frequency of this signal?

 

Hi
I have a differential signal which will be transmitted from a system (transmits and receives the data through 2 wires only, there is no GND wire).The differential signal has 3 voltage levels -12V , 0V and +12V. I have measured this differential signal with the help of oscilloscope by keeping one wire to GND of oscilloscope and one wire to positive of Oscilloscope.
The differential signal was shown below.

I need to decode this data i.e, convert to single ended signal but here the issue is when this differential is converted to single ended signal using AM26LS31CDBR IC, 0V logic level is either converted to High or low, but it is not following the same logic i.e. High or low.
I need a transceiver which converts this three state differential signal to single ended signal and it also should convert the same converted single ended signal to differential which should be same as three state differential input signal .

Can anyone suggest me the solution for reconstructing the differential signal.


View attachment 226802

Time period is 32microsecs

 

 

hi LT,
Your description is not very clear, you say that you are using a Differential Line driver to receive the diff signal and convert to single ended.
Please confirm.
E

 

yes

 

Which seems to mean the bit width is about 66 kHz if your sketch is to scale.

Is the signal subject to much common mode noise?

What are you using to convey the signal between the place it is generated and the place you want to decode it?

 

Hi LT,
How can the +/-12v Signals be connected as inputs to the 26ls31.?
E

 

If the source has a low enough output impedance, you could use a fast opto-coupler, or two of them back-to-back if you also need to detect the Zero Volts state.

 

The two wires from the system are directly connected to Y & Z pins of IC

 

The two wires from the system are directly connected to Y & Z pins of IC

hi,
That means you are using the 26LS31 as Line Driver NOT a receiver.

I would suggest a sketch showing how the system is connected.

E

 

Take step back

you say you have a three level signal in ,
+ve, gnd and -ve.

If so its impossible to convert that to a single signal,
the input has three states, where as the single signal only has two states.
if you do want to keep the three states of the input, you need at least two signals out.

Are you certain the 0v is relevant ?
it looks to me as if you just have a "clock" there, there is no information contained in each individual bit,
It looks more like a timing signal used in some radio such as MSF,
its the presence of the "clock" and for how long its present that matters, not that its +ve, gnd or -ve.

Can you clarify where the signal originates from and the encoding upon it,
as the decoding could be a lot simpler than needing to detect all three states.

 

I think it is Manchester (Bipolar RZ) encoded data which contains data that control a system and receives the data (acknowledgement from the system)
I am not sure about the encoding technique.

 

Why not use an op-amp as a difference amplifier, which is precisely a differential to single-ended device? Something like this:



With R1=R3 and R2=R4, then \[V_{\text{OUT}} = (V_1 - V_2) \frac{R_4}{R_3}\] This has the advantage of preserving the common-mode noise rejection of the differential signal, while also letting you scale the output range.

 

If it is bipolar RTZ, then all three levels are used.

What makes you think this is a differential signal ?
From your scope measurements, it could just as easily be single ended.

Do you have the circuit of the spec of the transmitting device ?

 

I need to convert this differential signal to single ended. I have used AM26LS32ACPW receiver to convert this differential to single ended signal but the 0V level is converted to either high or low based on the transition, but in datasheet for that voltage level the output is indeterminate state, i cannot understand that indeterminate state means.
please suggest me a line receiver that can converts this 3 state differential signal to single ended which should represents the 3 state.

 

hi LT,
You cannot use the 26ls32 for a 3 level data detection.
It is designed to be used as a differential receiver, see this application PDF.

BTW: indeterminate means it could be a High or Low level..

E

 

I need to convert this differential signal to single ended. I have used AM26LS32ACPW receiver to convert this differential to single ended signal but the 0V level is converted to either high or low based on the transition, but in datasheet for that voltage level the output is indeterminate state, i cannot understand that indeterminate state means.
please suggest me a line receiver that can converts this 3 state differential signal to single ended which should represents the 3 state.

Please,
read what we have been saying to you
don't just re state the question,


We have all given you suggestions on what we need to know to help you further, and reasons why your proposed solution is not working,

 

How do you want the ground state to be represented in the output signal? Idea: Have a separate ground state binary output or recreate the signal as a ternary signal all above ground?

 

Do you want the ground state to be converted to 1/2 the high level?

Is the signal truly differential (two wires with neither going to ground)?
Just because a signal goes plus and minus doesn't make it "differential".