Optosplitter Project

DMXpro

DMXpro

Active Member
Hello all,

No questions from me today, I would like to take this time to share a project I've just completed for a course in my collegiate Electrical Engineering program. I built my own DMX512 optosplitter, with a custom 3-D printed enclosure. Link to my website:
ELT2720: Electrical Project's home page

This was tremendous fun to do, and I love sharing my work and interests with y'all.

Enjoy!
 
DMXpro said:
Hello all,

No questions from me today, I would like to take this time to share a project I've just completed for a course in my collegiate Electrical Engineering program. I built my own DMX512 optosplitter, with a custom 3-D printed enclosure. Link to my website:
ELT2720: Electrical Project's home page

This was tremendous fun to do, and I love sharing my work and interests with y'all.

Enjoy!
Click to expand...

Just one question: Are the outputs optically isolated from each other as well as from the input? That configuration is by far the most desirable in terms of system robustness. However, it requires a separate supply or DC-DC converter for each output, so it adds cost.

ST
 
Steve beat me to it. Some things to be aware of....

- While the device is isolated from input to output, you might wish to look at isolation between outputs. That way, if a device fails on one output, it doesn't take out the other outputs.
- If the data is isolated, it is likely desireable to isolate the power between outputs/input. If a failure hits the device and finds its way to the power supply common, if there is only one supply, then all outputs/input can be damaged.
- Use cautoin with a 6N137 as the receiving device as it it has a very high current requirement to drive the opto LED. You may not be able to place more than one device on the same DMX line with the input to the splitter due to the heavy current draw of the 137. I think you will also lose the benefit of a differential data circuit since the LED really isn't doing the necessary comparator work.

Please understand that I am not knocking the design since I can tell you worked hard on it and I can tell you that it will certainly work. I think it's cool to have peer reviews, and in this way, you can take these comments as you wish and decide if any of this matters to you. If so, then you can do the research to see how to solve these problems. You can also PM me if you would like to know more.

I designed and built several interface products in college and am THRILLED you are doing the same!!!!

Great work here!!!

David
 
I have considered some design issues, I knew the individual isolation was popular. What confused me at first was the opto's seemingly single-ended output, which the DBT chips work with to produce a differential output (or so it would seem...). I am currently refining the design in my own time, it was great to get started with the current board. Also, this circuit probably works 5x better "in a pinch" than a Y-adapter, so there's that. After several extensive nights of reading datasheets on both the opto and the rs-485 DBT, I decided the design would work, and would certainly get me a grade. :)
I really appreciate the feedback, David, it means more than you know.

Regards,

DMXPro
 
STEVETERRY said:
Just one question: Are the outputs optically isolated from each other as well as from the input? That configuration is by far the most desirable in terms of system robustness. However, it requires a separate supply or DC-DC converter for each output, so it adds cost.

ST
Click to expand...

If you'll see my reply to David North, the current design does not isolate between each output. It is on the list of things I'd like to improve in my revision of this circuit. And, for personal use, I'm okay with it having only one isolation point (at the least, it's a good proof of concept).
Thanks for your question!
 
On the user end of things -- I understand this was made for your own purposes, and not as a commercial product; but still -- it'd be nice to have it either rack-mountable, or able to be clamped to a pipe (or, best of both worlds, both), which is how it'd be used in the wild.

The 9VDC power supply also seems a little weird; I'd rather have something that runs off of mains, and would therefore use either a IEC connector, or PowerCon.

Kudos for sticking with 5-pin. I think if it were 1/4" and 3-pin XLR jacks, it might be mistaked for a DI box. ;)
 
theatricalmatt said:
On the user end of things -- I understand this was made for your own purposes, and not as a commercial product; but still -- it'd be nice to have it either rack-mountable, or able to be clamped to a pipe (or, best of both worlds, both), which is how it'd be used in the wild.

The 9VDC power supply also seems a little weird; I'd rather have something that runs off of mains, and would therefore use either a IEC connector, or PowerCon.

Kudos for sticking with 5-pin. I think if it were 1/4" and 3-pin XLR jacks, it might be mistaked for a DI box. ;)
Click to expand...
9VDC because I wanted to keep it compact. I thought of drilling into the PLA and add an eyelet or a hole for a pipe clamp.
 
TS or TRS is not the best choice for power; the plug shorts out when it is inserted. There are locking (threaded) coax power connectors available, mount a screw eye to the case to use as strain relief, or hard-wire the power brick.

One issue with the schematic: You have all of the enable pins of the 75176 tied together, but they are not driven or pulled to a valid logic level. I'd pull them up to Vcc with a 1k or so resistor; inputs should not be allowed to float.

To minimize loading from the optoisolator, I have used a 75176 as the receiver and optoisolated its output in the past. For power isolation, you can get some small (1W) isolated DC-DC converters for under $3 (Recom's ROE-0505S should work)

/mike
 
Last edited:
n1ist said:
TS or TRS is not the best choice for power; the plug shorts out when it is inserted. There are locking (threaded) coax power connectors available, mount a screw eye to the case to use as strain relief, or hard-wire the power brick.

One issue with the schematic: You have all of the enable pins of the 75176 tied together, but they are not driven or pulled to a valid logic level. I'd pull them up to Vcc with a 1k or so resistor; inputs should not be allowed to float.

To minimize loading from the optoisolator, I have used a 75176 as the receiver and optoisolated its output in the past. For power isolation, you can get some small (1W) isolated DC-DC converters for under $3 (Recom's ROE-0505S should work)

/mike
Click to expand...
I used a schottkey diode on the input jack to protect against reverse voltage. Also, many of the schematics are older than the final product: The enable pins are pulled high with a 4.7KΩ resistor in my current design. This wasn't included in the schematic I capped for the website. This design, although criticized, was a basic proof of concept and I am refining it.

Thanks!
 
Excellent work! I see profit margin in your future.
 
Nice work on the project! DC wall warts tend to fail prematurely due to internal heat cooking the filter capacitor. A better design is to use an AC output wall wart and then put the rectification, filtering, and regulation inside the equipment where it is cool. Not only will it last much longer, but it also puts those components where they can be accessed for repair.

The DC supply choice makes perfect sense to get the project done quickly and more easily, like for a class. But, please keep this in mind when you start designing products for sale.
 
And if you design it properly for an low voltage AC supply, it will work with a DC supply of either polarity, so it won't go bang when someone plugs in the wrong wallwart...
/mike
 
People! This was a project for a class in school! There was some good feedback above, but seriously, UL listing? give me a break. This was not a project on building a power supply, why spend double the time building a supply that is not part of the grade(although if you did use an internal supply, power-con would be a choice connector.)

As to building a supply that can run on dimmers, frankly that is a poor idea, for many reasons; think about it.
 
Project looks great. I like the look of the enclosure. It looks better than most of the cheap splitters on the market. It's been a while since my circuit design classes so I'll pass on commenting on the circuit or PCB design. The biggest critique for me would be the use of the 1/4" TS connector for power. That's just not a very good design. Anything that would possibly short your power supply is not a good idea.
 
I know its sloppy, and obviously not UL listed :p but I have gutted a wall wart, and placed it inside an an enclosure, so I could use a regular edison power cable. I needed 120v ac and 12vdc, so it made sense to eliminate plugging 2 things in. This was for personal home theater project, but it does clean up the look and ease of use.
 
You must log in or register to reply here.

Similar threads

L
The Remarkable Journey of DyLOS
Replies
0
Views
722
Larry Beck
L
Newsletter
Highlighted posts from Wednesday, Sep. 07, 2022
Replies
0
Views
896
Newsletter
Newsletter
gafftaper
Design The New Pathway eLink and my lighting system upgrade
Replies
1
Views
1K
tyler.martin
T
Newsletter
Highlighted posts from Wednesday, Jul. 20, 2022
Replies
0
Views
729
Newsletter
Newsletter
Newsletter
Highlighted posts from Wednesday, Apr. 27, 2022
Replies
0
Views
694
Newsletter
Newsletter

Users who are viewing this thread

Total: 1 (members: 0, guests: 1)
Back