On Sunday, July 7, 2019 at 2:16:22 PM UTC-4, Charles Steinkuehler wrote:
> There's good info in the data sheet for that part: 
> http://www.ti.com/lit/gpn/thvd1451 
> ...which is an RS-485 transceiver (RS-485 is the same electrically as 
> RS-422, but the driver can be switched off). 
> RS-485 is a "multi-drop" protocol, and supports multiple transceivers. 
> Typically, the end nodes will have termination resistors populated 
> and the devices in the middle on the cable will have the termination 
> disabled.  See figure 32 (page 22) in the data sheet. 
> PCW said it's straight RS-422, not RS-485.  That means you only have 
> two devices making each device an end point, so add termination to 
> both ends. 


That's kind of unfortunate because you'd need a separate channel for each 
device. Now my head is spinning again because my encoders and stepgens are 
differential recievers/drivers so I could technically pretty much just 
repurpose one of each and get a couple of channels out of one set of 
connectors. I didn't put any impedance control or TVS on them but now I'm 
thinking if I make some accommodations I can just sacrifice encoders and 
stepgens for serial channels.


The ESD protection is application dependent, but you might also want 
> a small value series resistor between the THVD1451 and the cable (see 
> Figure 38, page 27 along with the recommend parts list) and see the 
> layout recommendations on page 29. 

 

> ...but don't sweat the details too much.  For a short range (a few 
> meters) point-to-point connection, the ESD and even the termination 
> resistors are not super important.  I'd still add them (it's cheap 
> insurance!), but I don't expect you're designing for 1.5 km long 
> cables that need to withstand nearby lightning strikes.  :) 
> That said, spindle motors can throw off enough ESD to cause problems 
> even with differential signals, which is why you want _something_. 


I just whipped the drawing up this morning, I was originally going to use a 
different chip and had a different drawing. Good to know it's on the right 
track though.

tvs and esd are somewhat of a concern since while I didn't really intend 
it, I can pretty much replace the x86 PC and 7i76e on my mill with this. In 
that case I use an 8i20 for my spindle with a power supply that consists of 
some big caps, a couple of SSRs and a bridge rectifier. There's 240v AC 
going into that box 320v DC coming out. While I shielded all cables, a 
little protection on the data lines is probably a good idea. Not sure of 
the performance of the DE10 Nano itself just yet since I just load up an 
Axis config and fire up halshow for the time being. Unless I converted my 
mill I probably wouldn't have an actual machine to run on this any time 
soon.

Every time I think I'm damn near done with it I come up with something else 
to complicate things lol.

-- 
website: http://www.machinekit.io blog: http://blog.machinekit.io github: 
https://github.com/machinekit
--- 
You received this message because you are subscribed to the Google Groups 
"Machinekit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
Visit this group at https://groups.google.com/group/machinekit.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/machinekit/a2e4e252-eda6-41a6-9dfe-971499328d56%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to