Dodge Cummins Diesel Forum banner

1 - 8 of 8 Posts

·
Registered
Joined
·
737 Posts
Discussion Starter #1
I am currently using 1 OBD II splitter to run my AMP steps and my switch for tune adjustment. I am looking to add and EDGE CTS3 Insight which means I need another splitter. So, how many splitters is too many? Can I get away with using 2 without issues?

I've heard the AMPs can be hard wired however I have no idea how to do this. I'd be interested to learn.
 

·
Registered
Joined
·
737 Posts
Discussion Starter #3
🤣
One of the things I'm trying to avoid! I've got an annoying amount of accessory wiring running stem to stern already and I will be adding more. It's getting to be a real nightmare.
 

·
Premium Member
Joined
·
1,258 Posts
Maybe look into using the edge as your sotf switch and a monitor. I had seen that's possible now. Personally one split is all I recommend. Amp should have a different wire harness if you want to hard wire them also
 

·
Premium Member
Joined
·
1,005 Posts
It appears that all the OBD II splitters and pass-throughs I've just looked up are purely passive devices, so they're not repeaters -- just cable splitters.

The use of a repeater on a network generally reduces bandwidth, since there's a delay in anything getting through the repeater that slows down anything accessing the network through the repeater. A cable splitter on the other hand won't affect the network unless there's so much extra conductor that the cable itself starts to be felt back to the network as an impedance. That shouldn't be an issue until you start adding devices several tens of meters distant to the network controller, or dozens and dozens of splitters. The connections themselves will add some impedance of course, but generally the equivalent impedance for a well engineered plugged in connector is quite tiny.

OBDII supports three separate networks: CANbus, J1850, and ISO9141. It appears that Ram uses the ISO9141 network for some or all devices. I can't comment to that, except that I know my truck has seen network comm issues randomly for no good reason. Being a LL, it has tons of accessories, so I believe either the 9141 or CANbus network -- whichever one is used more heavily -- must be near capacity, or that there are participants that hog bandwidth. It appears that J1850 is an old standard and not used much at all. Or so I read. CANbus I can comment on from personal experience. It's limited to a couple hundred devices. Any more than that and some of the participants will start to get confused over timing elements if most of the devices are active participants -- if they're sending and receiving data.

Now, if I'm reading accessory descriptions correctly, both the AMP steps and the Edge Insight are entirely plug and play. Although the AMP steps will work with a pass-through harness that's unique for RAM, which should tell us something. You hook them up and they work. This would indicate that they're consumers of information on the existing network, but not active participants. I mean... Ram doesn't code their networks to support the steps or the monitor. They just work, on their own. That tells me they're monitoring network traffic messages that are already there. Now, supposing that both the steps and the Insight are compliant network interfaces -- which is to say, they don't load down the network -- then we're only talking about adding two network nodes. The physical splitters don't count for anything -- only the number of nodes we're adding. And you're saying two, which shouldn't be any issue at all, since they're passive nodes that won't add to the network traffic. Probably.

So given what I know about how the network nodes work and how the splitters work, I wouldn't expect any issues at all.

But it would still be great to hear from someone who might have actually done it. I could be wrong.
 

·
Registered
Joined
·
2,039 Posts
Maybe look into using the edge as your sotf switch and a monitor. I had seen that's possible now.
Don't quote me but I'm pretty sure the Edge SOTF function only works with CCV type switching, not OBD port so it probably won't work in his application..
 
  • Like
Reactions: RamHunter18

·
Registered
Joined
·
737 Posts
Discussion Starter #8 (Edited)
It appears that all the OBD II splitters and pass-throughs I've just looked up are purely passive devices, so they're not repeaters -- just cable splitters.

The use of a repeater on a network generally reduces bandwidth, since there's a delay in anything getting through the repeater that slows down anything accessing the network through the repeater. A cable splitter on the other hand won't affect the network unless there's so much extra conductor that the cable itself starts to be felt back to the network as an impedance. That shouldn't be an issue until you start adding devices several tens of meters distant to the network controller, or dozens and dozens of splitters. The connections themselves will add some impedance of course, but generally the equivalent impedance for a well engineered plugged in connector is quite tiny.

OBDII supports three separate networks: CANbus, J1850, and ISO9141. It appears that Ram uses the ISO9141 network for some or all devices. I can't comment to that, except that I know my truck has seen network comm issues randomly for no good reason. Being a LL, it has tons of accessories, so I believe either the 9141 or CANbus network -- whichever one is used more heavily -- must be near capacity, or that there are participants that hog bandwidth. It appears that J1850 is an old standard and not used much at all. Or so I read. CANbus I can comment on from personal experience. It's limited to a couple hundred devices. Any more than that and some of the participants will start to get confused over timing elements if most of the devices are active participants -- if they're sending and receiving data.

Now, if I'm reading accessory descriptions correctly, both the AMP steps and the Edge Insight are entirely plug and play. Although the AMP steps will work with a pass-through harness that's unique for RAM, which should tell us something. You hook them up and they work. This would indicate that they're consumers of information on the existing network, but not active participants. I mean... Ram doesn't code their networks to support the steps or the monitor. They just work, on their own. That tells me they're monitoring network traffic messages that are already there. Now, supposing that both the steps and the Insight are compliant network interfaces -- which is to say, they don't load down the network -- then we're only talking about adding two network nodes. The physical splitters don't count for anything -- only the number of nodes we're adding. And you're saying two, which shouldn't be any issue at all, since they're passive nodes that won't add to the network traffic. Probably.

So given what I know about how the network nodes work and how the splitters work, I wouldn't expect any issues at all.

But it would still be great to hear from someone who might have actually done it. I could be wrong.
Well that was certainly a read. I think I even understood the bulk of it! Basically, what I got is I'll be ok to add another split for the CTS3 since it's generally passive as well. A 2nd splitter will allow me to run my accessory items and still have a port open to run my code reader and such without unplugging something. Currently, the AMPs get unplugged when I'm reading codes/running AlfaOBD.

I can say for certain that the AMP plug is passive only. From reading posts here, AMP themselves say that the plug only "listens" for the door open signal to deploy the step.

The next thing I'm gonna run into is where to put the bank of EAS sensor connections I'll end up with and exactly how many I can have. Adding a pyro is the first of at least 4 if not 5 other sensors for things I want to monitor. That's topics for other threads tho.
 
1 - 8 of 8 Posts
Top