Chauvet Obey 50

[PantoDame]

I have been using a Chauvet Obey 50 to control some Par 56 LEDs. There is a 3 pin din output and a 5 pin din output. I have been using the 3 pin. I have some 500w spotlights that are controlled by a DMX dimmer desktop and dimmer racks. I have found that I can plug the 5 pin din into the Obey and it controls the older spotlights. Can anyone comment on if this is an acceptable way of doing it. And if so should I use an end of line resistor at the dimmer racks end as well as on last led. Thanks

[sandall]

To be pedantic, these will (should) be XLR, rather than DIN sockets, but to answer your question, yes you should be able to use both at the same time. Your dimmer rack should have a termination resister built in (though there may be a switch or jumper involved). The purist answer for the leds is yes, terminate the last one with a 110(-ish) Ohms resistor, but they will probably be fine without.

[sunray]

I didn't think there is an Obey 50 as they used 50 for their 'Stage Designer 50'

 

on those 2 XLR sockets are simply connected in parallel and there are many mentions on this site to never use branches on DMX.

[DrV]

1 hour ago, sunray said:

on those 2 XLR sockets are simply connected in parallel and there are many mentions on this site to never use branches on DMX.

But there aren't any branches if you just have one line plugged into each XLR - it's a single line with the desk in the middle so you should have a terminator at each end. However for relatively short lines where the propagation delay from one end to the other is small compared with the bit period it's not terribly important.

There are people on here with a much deeper understanding of transmission line theory than I but I think I'm broadly correct.

Dave 

[themadhippy]

Quote

it's a single line with the desk in the middle so you should have a terminator at each end.

But  what value terminator as theres now 2 of em in the circuit, is it a simple parallel connection so 240 ohm at each end or is  there more to it?

[sunray]

2 hours ago, DrV said:

But there aren't any branches if you just have one line plugged into each XLR - it's a single line with the desk in the middle so you should have a terminator at each end. However for relatively short lines where the propagation delay from one end to the other is small compared with the bit period it's not terribly important.

There are people on here with a much deeper understanding of transmission line theory than I but I think I'm broadly correct.

Dave 

I wouldn't claim to be expert either, as I understand it this means it is terminated in the middle at 120Ω (or whatever the current fashion is) with a line in each direction which should both be terminated resulting in something possibly near 60Ω depending different distances and multiple reflections from both directions. Most transmitters designed to be anywhere on the line would tend to be tri-state. 

I do tend to agree the low working frequency could make this less likely to be a problem however I have discovered how easily data transmission can get corrupted by incorrect assembly. It's another of those try it and see situations which will be fine... until it isn't just before a show starts.

One wouldn't dream of fitting a Tee on a radio transmitter and a couple of random lengths of 50Ω coax feeding a 50Ω aerial each. But in that situation there is risk of damaging the output device due to the bad match.

 

I've always wondered why they didn't fit a second 20 pence driver chip when there are 2 sockets to simply provide 2 outputs and something I've thought of doing in my desks.

[DrV]

No, it shouldn't be terminated anywhere else except the ends. That's what "terminate" means. See page 14 of the TI 75176 datasheet for a picture.

[indyld]

People get bent out shape about DMX termination on the internet, but oddly you can travel far and wide and evidence would suggest that a significant portion of venues, tours, theatres, and one offs all over the world can't show you a terminator. 

Doug Fleenor has/had a page on why you should terminate. 

As suggested, I imagine the 5 and 3 are just common inside the desk. DMX will appear to work if you use a fixture with 3 and 5 pin connections as a Y split, but it's not recommend. 

If the desk spec only has one universe, then technically you can't send two lines straight out. I can't teach students to do that, for example. 

However, when you try it no doubt appears to work. 

 

Edited November 15, 2023 by indyld

[DrV]

Yes, people do get hung up about termination of DMX - it's only 250kbaud over (usually) relatively short distances. However, just because something usually works doesn't mean that one shouldn't try to use the best practice from an engineering point of view.

RS485 is designed as a multi-drop bus. Usually we drive it from one end and receive everywhere else so it never occurs to us that we could legitimately drive it from anywhere along the bus. Take an RDM fixture responding to a status request; that's sending a signal just like the desk does, with all the same physical layer characteristics. The signal goes off in both directions and anything on the bus that is interested can take note.

Using a fixture as a y-split is not the same at all and would definitely be poor practice. Will it work? Usually yes but in the limiting case where you're running very long lines with transceivers which are at the very edge of their tolerance then maybe not.

Engineering is all about reducing the potential for weaknesses at every possible step so that on the rare occasions when all the tolerances gang up against you the system as a whole continues to perform.

[sleah]

This just crossed my mind.... how is connecting to both the 3pin and 5pin outlets of a desk (in effect making it inline rather than one end) different to the normal practice of looping fixtures using the male and female connections? Electrically it's identical (taking termination that may or may not be in the desk out of the equation for a moment).

The obvious difference being the desk is generating data where the fixtures are receiving data, but other than that...?

[PantoDame]

Thanks for all the responses. I did mean to say XLR not din and the desk is a Chauvet Stage designer 50. 
there is  a 120ohm resistor in t  last LED but nothing added on the dimmer racks. I don’t know if there is one internally. all I can say is - it works , but is that a good enough comment to continually using it.

[sandall]

1 hour ago, PantoDame said:

.... is that a good enough comment to continually using it.

Yes🙂

[sunray]

19 hours ago, DrV said:

No, it shouldn't be terminated anywhere else except the ends. That's what "terminate" means. See page 14 of the TI 75176 datasheet for a picture.

Yes that is the correct method of terminating an RS485 line, the issue here is the desk is built as one end of the line and includes a 120Ω terminated driver, hence the 'requirement' to terminate the other end with 120Ω.

Doing a Y split at the desk is no different to doing a Y split at any other location in the system.

As has been stated it is very likely to work successfully on smaller systems and I confess my systems are not normally terminated simply due to them being designed on the hoof and getting the final fitting up and focussed brings an end to ladder work.

 

I have had problems, on one occasion I inadvertently introduced a branch (a fitting with both 3 and 5pin connectors) of around 2m to a single fitting and it floored the section from a splitter

That whole section was only about half dozen fittings and 15m of cable max.

Edited November 16, 2023 by sunray