Communication Signal Help

[r_olson_06]

I am working on a communication signal for a fire suppression system. The signal is 24VDC circuit. The pulses on a oscilloscope will only dip to a low of +18VDC. So a pulse height of only 6VDC. Is this common and how is this possible to dip only 6 volts and not the full 24volts? If so what kind of communication style is this? It is hooked up as a 2wire daisy chain similar to a RS485 loop or profibus. Any insight would be great.

 

[plow]

What circuit are you reading? What system? What are you trying to accomplish? Is it conventional or addressable?

 

[kaffine]

I am not familiar with RS485 or profibus so this may not be valid.

One of the modules on the daisy chain has failed and supplying constant power to the chain. When another module tries to transmit data it can not sink enough current to bring the voltage level down to 0. Could also be there is another signal shorted to the bus somewhere supplying power. Could try and disconnecting modules one by one to see if the problem goes away.

The other thing is does 485 or profibus require terminating resistors? Perhaps one or more of the resistors is missing or failed.

Is RS485 or profibus a single ended signal or a differential signal? If differential are you comparing the signals or are you referencing a ground?

Again I said I am not familiar with rs-485 or profibus so I am just assuming the values are wrong and it should be dropping the voltage down to 0V

 

[ironmutt]

i might be wrong but i am pretty sure if your just reading voltage you may not have the right osilliscope . if the signal is pulsing fast enough it will look like only a slight drop in volts even though it is going full off and on it is so fast the machine cant read it

 

[plow]

It sounds like he may be reading supervisory current. That is IF I'm "on the same wave length" as him (pun intended).

 

[justsam]

I am not familiar with these systems at all, so all I can offer is to be sure you are AC coupled to the signal and you are not looking at any DC sealing potential that may be present, and not allowing you to see the full voltage swing. Direct probe or 10x probe?

 

[rockwithjason]

you can engineer a signal level to be anything you want. these protocols with fire alarms and suppression are often proprietary to the manufacturer. can they offer any insight on this?

 

[r_olson_06]

Called manufacture on the signal and found out very little on the signal. With a noise level running about 10VAC (peak to peak) for a duration of 10ms and re-ocurring every 200-250ms sound like a noise issue. Also with the manufacture saying this signal is really prone to noise installed on a system with 20+ Servo Drives. Sounds like Trouble.

 

[plow]

By code it has to be UL listed. If something bad happens, Its yo baby. DACT units are cheap. Good luck.

Phillip out

 

[rockwithjason]

if it occurs regularly try to find the source of the noise by turning off equipment one at a time until the noise clears. you may need to isolate the cable in piple and use shielded cabling to null out the noise.

 

[Norcal]

By code it has to be UL listed. If something bad happens, Its yo baby. DACT units are cheap. Good luck.

Phillip out

A code cannot require UL or other specific listing, they merely require listing which is understood as listing by a NRTL, Nationally Recognized Testing Laboratory which UL is one. Requiring a specific NTRL by a code means lawsuit time.

 

[Rob_b]

quick question...what type of cable are you using for this RS485/profibus line? It should be a twisted shielded pair(beldon) with the shield landed on ground at one end only as a drain. You can't run this cable with any other current carrying conductors in a conduit. I work with RS485/DH+ every day and it can be a PITA to track noise.