• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 260 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

AM26C32: Making a Universal Differential controller

Peter Lewis
Prodigy 10 points
Part Number: AM26C32

Hi,

I need to make a universal Differential receiver module.

The specification is that I am using CAT5  and I have 4 rows of SPI LED 800khz max throughput Each row separated by 1m.

When I am sending from the LED SPI controller I get interference greater than 10m . In some cased I need to go up to 30m

So in searching the web , I noticed that some companies used differential drivers. I purchased a whole lot of the AM26C32CN (25 pcs) and AM26C31CN(5) .

the idea is to make one design for the receivers so that I can produce in volume the same item and use it for the 4 outputs. Each output is 1m apart.

The module design I thought would have the opportunity for all 4 outputs of the receiver which is battery powered by 2 x 3.7v Batteries and using a 5 volt regulator

Then the 4th module in the chain would have the termination resistors avail on a dip switch as the requirement of being the last Receiver in the chain being terminated.

Each module will have Ethernet in and bridged to Ethernet out with a tap to the receiver chip.

Before I manufacture the samples of this item and design the board with the strict rules for the differential lines as per the datasheet. I wanted to make sure this circuit was ok of should there be any other design changes. I do not know what the Value of G and (NOT G) should be . I suspect the G should be +5v and (Not G) not connected.

Any assistance and guidance on this design would be greatly appreciated. 

I manufacture LED Dancefloors in South Africa which this is what I require to make the floors run correctly without corruption on distances further than 10m 

Thank youUniversal Differential Receiver Circuit.pdf

 

  • +1
    Guru 316690 points

    The termination looks OK. (The termination resistor must be at the end of the bus, so if there is a distance between the resistor and the receiver, having the receiver before the termination is correct.)

    The G/G inputs must not be left floating. See table 1 in section 8.4.1; if you never want to disable the receivers, connect G to VCC and /G to GND.

  • +1
    TI__Guru 52172 points

    Hi Peter,

    A couple notes:

    1. The termination looks fine in how you placed it - the value of the terminations I don't see - but they should be 100 Ohms (they can be greater than 100 Ohms - but this is a RS-422 part - which is spec'd w.r.t. 100 Ohms). 

    2. G should be connected to VCC and /G to GND to keep the device always enabled - or you can connect a GPIO from a controller to enable/disable the device when certain conditions are met. 

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    Prodigy 10 points

    Thank you Parker for this info. 

    From the truth table it looked like G was suppose to go to +ve and /G to gnd but looking further down in the table it looked like if one was enabled the other did not care. 

    Yes i was thinking of either a 100 ohm or 120 ohm Resistor pack for the termination. Thank you for the advise

  • +1 in reply to Peter Lewis
    TI__Guru 52172 points

    Hi Peter,

    Yes technically its an OR gate - so only one of the pins needs to be controlled to control the device. Its just best practice to never leave logic pins floating - but it shouldn't be an issue if you just attach VCC to G for always enabled operation. 

    100 or 120 Ohm terminations are both okay - so no major concern here - it is best to match the termination to the characteristic impedance of the transmission line of the differential bus to mitigate reflections - but based on your signal speed (800kHz), max bus distance (30m), and estimated signal velocity (78% of c - this could change but generally a lot of cables used in RS-485/RS-422 should be pretty close to 78% c)  would put 1/10th of the signals wavelength (for fundamental frequency of pulse train) around 29.2m - so most of the energy that realistically could be subject to reflecting in the first place is going to be in the higher harmonics and not really fundamental - so slight differences between characteristic impedance of transmission line and termination really shouldn't really cause any issue. 

    Best,

    Parker Dodson