MCU circuit in a fishing vessel?

[UroBoros]

Hai

I have a PIC based circuit to be deployed in a fishing vessel.

Supply is from 24 volt battery connected with a charger from altranator in the vessel.

The engine almost 2000 rpm in boat.

I am trying to design a spike and transcient free power supply for the system.

I have tested the circuit with a 24 volts rectified power supply which I first drop to 14 volts using a LM317 and then distribut within different boards in the unit placed meters apart. communication is via RS485 and locally uses 7805 to drop to 5 volts.

can anybody suggest a suitable inductor capacitor arrangement to be placed in the input supply rail from the boats power supply bus?
Any application notes or literature for this?

Torroid will be preferable as it is small in sixe so that it fits in PCB.

Thanks
picstudent

 

[banjo]

The LM317 is good up to a 40V differential between its input and output voltage. The alternator spikes should never exceed this. I really do not think you need to use an inductor to limit power spikes. The LM317 should regulate them out. That is what its job is.

If you are having problems, I would be more concerned with ground circuit. Since you have multiple boards and the boat can have large currents on the ground from a variety of other devices, each board's ground can be at a different relative voltage.

When communicating via RS-485 remember that a ground return is required. This is not a two wire system that many people believe it is. Between the boards, should be a ground for just the RS-485 communication stuff.

Finally, if you really want to filter the 24V, research commerical filters used on boats. I bet they make them. I know they make filters for automobiles that reduce alternator noise getting into the radio.

 

[jhbbunch]

Do not use an lm317. Use an automotive grade LM29XX
voltage regulator or equivalent.

 

[UroBoros]

Hai

Thanks for the reply.

Which is the one in LM29 series which can handle upto 1 amps?

Also RS 485 requires common ground?

Actually my diiferent boards i power through a bridge to prevent problems from plarity reversal by accident .So the system grounds are not common. They differ by .75 volts. But So far works ok in RS485.

Will this create problems? . I am not sharing common ground with vessel.
I am taking DC24 volts at a RS485 node and run 4 core wire between systems.
rs485 two wires and supply 2 wires.

Coments please. System is not thoroughly field tested so far.


Thanks again
picstudent

 

[IanP]

0.75V difference shouldn't be a problem for RS-485 ..

Also, LM317 will not remove spikes from power generator and inductors on both rails (+V and 0V) should solve this problem ..
I would go for something like 100µH - 1mH, and then, of course, electrolytic (≈1000µF/1Adc) and ceramic (as usual 100nF) caps ..

Regards,
IanP

 

[UroBoros]

Hai

Thanks Ianp for the suggessions.

Here is what I have in mind
[the coil are wound on a 1 inch torroid ,some 6 turns.


I have seen this type is some transceivers supply

Also please clarify weather rs485 requires a common ground.

if we connect only rs 485 lines between two systems will it work?

picstudent

 

[IanP]

The RS-485 does NOT need common ground - it is differential transmission and grounds do not need to be connected ..
If you use TSP (twisted shielded pair) connect shield only at one end ..

BTW. Filter looks good, depending on the AL of the toriod, 6 turns may be not enough ..

Regrds,
IanP

 

[banjo]

Please refer to:

**broken link removed**

In this document you will find a reference to the needed ground in RS-485 systems. While many people will tell about their successful system that did not use a ground, I can tell you about damaged transceiver chips in systems that did not include the ground. Perhaps you will not see the problem on a boat. Where I experienced the problem was RS-485 communication between buildings. The cabling had to be run by union workers which were lazy. They gave me onto four conductors for my full duplex system. Without a common ground between the RS-485 chips, they had to rely on local earth ground which can have a large offset. We had repeated intermittent transciever failures, often after power problems or lightning storms. After replacing the chips, it would work for a few weeks and then fail again. After researching the transceiver chips and adding the ground wire, the problems went away. No more damaged chips because both sending and receiving devices were working from the same potential.

My suggestion would be to contact the manufacturer of your transciever chips and ask the grounding question very directly. Many manufacturers show simplified diagrams that do not include the ground.

Added after 8 minutes:

With regard to the bridge power supply, I would remove the 1N4007 diode. This is a crowbar diode that will blow the fuse when polarity is reversed. Since you are using a bridge rectifier and want to be able to run with either polarity, this diode only causes problems. Again, the diode's purpose is to protect downstream electronics from polarity reversal by blowing the fuse.

 

[UroBoros]

Hi
Thank you very much for the detailed explanation about ground issue.


Also thanks a lot for pointing out the diode issue.That diode if remained will compramise my intention in placing that bridge.
Thanks a lot

Picstudent