MAX488 unused inputs and outputs

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
How should I tie the unused transmit and receive stages of a MAX488 IC?

Capture.JPG

Instinct tells me that for the receive part of the chip, the two differential receive pins should be shorted (and not connected to either Vcc or gnd) and its output floating.
As for the transmit stage, the input could be connected to ground, and the two transmit pins left floating.

Am I right in these assumptions?
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
Half. 0 V between the two receive input pins is not an unambiguous state. I would tie one up and one down through resistors.

ak
Interesting... it makes sense. Then again, I'm not sure how the MAX488 works. Does it send a negative and a positive voltage through its outputs? If so, isn't it expecting something similar at its inputs?

Capture.JPG
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
It is a 'Differential' data, IOW just opposite polarity.
Is there any reason you are not using a transmit or receive only IC, if you are not using one of the RX/TX sides?
Is this for a single ended encoder by any chance?
Max.
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
It is a 'Differential' data, IOW just opposite polarity.
Is there any reason you are not using a transmit or receive only IC, if you are not using one of the RX/TX sides?
Is this for a single ended encoder by any chance?
Max.
Yes, it is for a two channel encoder. I have 30 of those chips lying around, so I thought I could put them to good use. Of course, I'll end up using a total of 4 of those suckers. Two to transmit from the encoder side, and two to receive at the MCU side.

So you agree with AnalogKid's suggestion then? Also, as future reference, what pair of Rx/Tx chips would you suggest for this application?
 

AnalogKid

Joined Aug 1, 2013
10,987
I'm not sure how the MAX488 works. Does it send a negative and a positive voltage through its outputs?
No. In round numbers, both outputs are 0 V to +5 V, 180 degrees out of phase with each other. One of the main differences between a RS-422/485 receiver input and a normal gate like a hex inverter is the enormous voltage range it can handle without failure. That range is TTL (0 V to +5 V) +/- another 7 V. But the actual signal range expected, independent of whatever ground loop DC it might be sitting on, is differential TTL.

ak
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
Ok then, one of the unused reciever's differential inputs will go to ground, and the other one to 5V. Both through 10k resistors. Sounds about right?
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
One last question. Twisted pair cable is very hard to find here, so I'll be using shielded cable instead. Cable length is of about 6m. Should I use two pairs of 2x22 shielded cable (one 2x22 cable for each channel), or will a single 4x22 shielded cable do? That is, transmit both channels through the same 4x22 shielded cable.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
Not sure if I posted this before but it is a good paper by Siemens on equi-potential bonding and also see ch6 on shielding and grounding of same.
Stick welders particularly, spread quite a large magnetic field.
Max.
 

Attachments

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
Not sure if I posted this before but it is a good paper by Siemens on equi-potential bonding and also see ch6 on shielding and grounding of same.
Stick welders particularly, spread quite a large magnetic field.
Max.
Excellent document, Max. Thanks for posting!
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,220
The 26LS31/32 have done good service for me in the past, but I try and use Differential output encoders to avoid using them if possible.
Max.
Your chips definitely look better than mine (and less expensive too). Yours can run at up to 10Mbps, while mine at only 250kbps. Not that it makes much difference for my purpose, since my encoder's maximum frequency is 100kHz anyway.
I'll buy a batch of those chips next time I shop in Digikey or Newark.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
Well... 99% of solving a problem is making the correct diagnosis... I'm building the circuit right now, and I'll let you know how it went.
Not totally relevant in this day and age, But I once maintained a couple of CNC machines that ran off of magnetic memory, every time a mechanic used a stick welder within 15ft, he wiped the memory, reloading was by paper tape, and took around 30 min.!
Max.
 
Top