Contact US

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

How to piggyback an AC ripple ontop of a DC power line

How to piggyback an AC ripple ontop of a DC power line

How to piggyback an AC ripple ontop of a DC power line

I am having this mad idea of coupling a carrier frequency on top of a 24Vdc power line. The idea is that a band pass filter will detect pulses of the ASK carrier and provide a PWM signal for application control but obviously only requiring 2 cables.
I know protocols such as DC Lin and DC Can are available and I was wondering what would be the simplest method (low component count) to provide something equivalent.
This is just to get the conversation going as I would really appreciate any input.
The above is a bit simplistic but fire away and I will reply expanding on my requirements, easier that way as this post could turn out to be a rival to 'War & Peace'

Many Thanks


RE: How to piggyback an AC ripple ontop of a DC power line

It's done regularly.

Look up X10.
Look up the Hart Protocol.
Carrier current communications.

It would likely be less optimal to come up with your own if you can use something already established, understood, and supported elsewhere.

I had the dubious pleasure of blowing a 460V breaker supplying  one thousand two hundred refrigerated containers while fielding a carrier current SCADA system when a 1,600V capacitor let go.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

RE: How to piggyback an AC ripple ontop of a DC power line

Some hotels retrofit high speed Internet into the rooms by installing their very own in-building ADSL system onto their existing telephone wiring. The point being that you can probably achieve several Mbps over a low voltage DC wire many thousands of feet long. 24Vdc is less than the voltages normally encountered on a normal telephone wire.

There are many other options, but the distance & speed high water mark is probably set by the latest ADSL technology.

RE: How to piggyback an AC ripple ontop of a DC power line

Thanks for the replies,

I haven't come across HART protocol which surprises me as I have done a fair bit of process control.

The 24Vdc supply is a power line providing up to 100watt for device drivers.

X10 doesn't really hit the mark as being AC transmitting at the zero level crossing in slow bursts provides a low baud rate. Also it doesn't support any method of auto configuration.

Lin is Master-Slave but only works with 16 nodes.
Can gets very heavy (fat) on footprint and MCU complexity.
Homeplug incorporates full tcp/ip with security.
Hart seems to be limited to 4 to 20ma process control.
I can't seem to get a spec on the new home plug C&C
There is cebus and various other PLC networks from broad band to narrow band applications and they are either to bloated with functionality or don't seem to provide what I require.

I am really interested how yamar are providing dc lin and dc can protocols over a 12Vdc - 48Vdc power bus but can't seem to find any technical data.

I really don't want to reinvent the wheel or a completely bespoke protocol to be honest.

What I want is a 64 node max peer to peer system which will be a sub bus to the previously mentioned, prob tcp/ip using upnp. Its main criteria is auto configuration in some sort of simple DHCP method with as high a baud rate as possible.
This then has to be squeezed into a tiny footprint as possible to run on low end MCU's with minimal component counts.

Ha thats all I want lol.

I have spent much time over various options and can't find a true fit unless you can expand where my oversights are wrong.

Still to rattle on ASK with a carrier frequency proving a signal to a usart seems about the simplest method as I am only trying to communicate over short distances 10-20m max.

Thanks for your input and I look forward to any further input.


RE: How to piggyback an AC ripple ontop of a DC power line

I wonder if power over ethernet protocol 802.3 a-f would suit your needs. I know a company in Bozeman, MT that has completely worked out such a network for LED giant screen video. They utilize FPGA and ethernet to network individual panels making up a single giant screen. The PoE has a limit of just under 13 watt but varied configuration can get as high as 72 watt. The data rate for PoE is as high as Gigabits/sec. The technology is exteremly mature.  

Best Regards

RE: How to piggyback an AC ripple ontop of a DC power line

That is a good suggestion xray.

I would be trying for that too.

PoE is a great idea even if you have customize some of the 'P' aspects.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

RE: How to piggyback an AC ripple ontop of a DC power line

Many Thanks again,

POE is a 6 wire with 48Vdc and due to cat5 is only really expected to power 'ultra lower power' devices due to cable lengths and all that (Thats from memory but its roughly right).

The nearest spec that is near to what I want is DC Lin or DC Can it an automotive spec for providing a vehicle DC power line 12DC to 48DC (2 wires) with comms along the same wires.

Apart from not being able to find much info from Yamar (the patent holder and semiconductor supplier) it still doesn't hit the mark as Lin specs are to limited and Can starts to become a little to fat.

Even though its looking like the DC-Can is the only spec out there that can provide something near what I require.

The other thing is if anyone could provide a little help here as I know it sounds stupid but honestly I am really thinking about creating a dedicated protocol ASK DC system.
I have previously reverse engineered a couple of Mitsi (PLC the programmable types) serial comms and already done quite a bit with TCP/IP.
Its the power electronics that I am a bit sketchy on and not the digital control protocol.

Anyway even though I haven't had the exact answers I want its really good to see your posts as they reflect my own thoughts and do help.

Many Thanks

RE: How to piggyback an AC ripple ontop of a DC power line

Cat5 is 8-wire, not 6, and your definition of "ultra-low power" is odd... I would hardly call 13W low power.

What are you trying to power that requires more than that?

Dan - Owner

RE: How to piggyback an AC ripple ontop of a DC power line

Sorry for not elaborating but from memory cat5 you have differential RX and differential TX and don't they use two other wires for DC supply?

Anyway I only want to use two wires Power Line Communication by imposing a carrier on a DC powerline.

CAT5 has 8 wires and 4 of them in most conditions are not used but still not what I require.

I also want a minimum of 100watt.

I have being doing some research and the closest almost brilliant system is AS-Interface but still one snag is its master slave organization and not peer to peer.

Thanks Anyway

RE: How to piggyback an AC ripple ontop of a DC power line

FWIW, in my opinion the best power-line networking out there is Lonworks.
go to "Products" then "Transceivers".
Basically made for mains use, they will also work on a low-voltage DC bus (an application note/reference design is available).
Bit rate is relatively low (around 5 kbps), but usually in control networks response time is much more important that datarate.

Best Regards,


RE: How to piggyback an AC ripple ontop of a DC power line

Let me try to summarize this mash-up.

You want 24vdc

You don't want the guaranteed response times of CAN.

You don't want a "fat" footprint.

You don't want single wire minimal LIN because it only works with 16 nodes.

You don't want a complex ADSL high rate 2-wire system.

For some reason HART won't fly..

You don't want to reinvent the wheel.

You DO want 64 nodes

You DO want DHCP

You DO want the highest baud rate possible.

You want this running on a minimal CPU.

You only want 20m max.

POE doesn't work as you want 100W.


You want 100W at these secret endpoints. That's 4A down 64 20m cables!

So you're going to have a 250A 24VDC power supply? This isn't looking too great.

You want to spend a month designing and shaking out some scratch, custom, ASK system and do a bunch of DHCP coding.

This means custom magnetics, and unknown RFI, EMI, thrashing.

What's the point?  Why don't you take a mature OTS technology and bend  it to your needs?  This is called leveraging, and is what good engineers do to get a product fielded with lowest risk and quickest development.

Ethernet uses almost NO foot print these days especially when you just use the UPNP part of the stack.

You discount Ethernet because it has 8-wires pfft!  It also has easily understood termination tools that are very inexpensive and reliable.  There is no cheaper cable due to its ubiquitousness.

You don't want LIN because it's only 16 channels?  So make a hub!  Use four nets.

Use what exists, is understood, inexpensive, and is adaptable.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members! Already a Member? Login


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close