SN75ALS181: RS485 network

The SN65HVD3088E is not an exact pin-to-pin replacement of the SN75ALS181, but if the A/B and X/Y pins were connected together, it will work.

Hi Yanagisawa-san,

I believe this is a duplicate of a different thread - please see answer here: https://e2e.ti.com/support/interface-group/interface/f/interface-forum/1327012/sn75als181-interface-forum and we can continue the discussion on that thread as it seems a little more active than this one. 

Best,

Parker Dodson. 

Hi Yanagisawa-san,

The SN75ALS181 is not rated to have 64 SN75ALS181's on 1 bus - the RS-485 standard allows 32 unit loads on 1 bus - the SN75ALS181 is rated to 1 Unit Load - so only 32 can be on a bus (and this number is lower if you have biasing resistors added to bus - by how much depends on biasing resistors).

The new suggested device SN65HVD3088E is rated to have up to 256 nodes on 1 bus because it is only 1/8 unit load (32/ (1/8) = 256).

If the bus has mixed devices (not all nodes have the same transceiver) then the limit is 32 Unit loads and each node's unit load rating needs to be summed and as long as the bus loading is <= 32 Unit loads it should be okay in terms of number of nodes. 

Best,

Parker Dodson

The RS-485 specification does not define any speed. In practice, skew adds jitter, and how much jitter you can tolerate depends on the receiving device.

The SN65HVD3088E has a better skew specification than the SN75ALS181 because it is designed for a higher data rate.

Hi Yanagisawa-san,

Clemens is correct - speed is not directly defined in the RS-485 standard - so there is no direct skew requirement. It should also be noted that RS-485 doesn't have a huge emphasis on precision, (its focus is robustness and speed (relative to long distance)) so if tight skew requirements are needed this may not be the best standard to pursue. 

As Clemens alluded to though the skew translates to jitter and much can be tolerated is really going to be a system level requirement. This is because the cable used in RS-485 can add jitter depending on the data-rate vs. cable length (and other cable properties). Faster devices will have smaller skews because the propagation delay is also smaller. 

Best,

Parker Dodson

Yanagisawa-san,

The value in datasheet is what we list as our max for tested conditions - we don't have different ratings than what are listed and we don' have a variance on it either -you need to assume max for the system when designing - anything less is insufficient and adds risk to the system - but once again in RS-485 systems  the cable is most likely going to be adding more jitter than the device in the vast majority of applications unless its a shorter bus - but then RS-485 may not be the best option. 

Best,

Parker Dodson