Skip to content

Instantly share code, notes, and snippets.

@darconeous
Last active November 8, 2024 16:20
Show Gist options
  • Save darconeous/8c7899c4d2f849b881d6c43be55066ee to your computer and use it in GitHub Desktop.
Save darconeous/8c7899c4d2f849b881d6c43be55066ee to your computer and use it in GitHub Desktop.
Hacking the Rectangular Starlink Dishy Cable
@WIMMPYIII
Copy link

So I have a couple of questions: Is there a good way to test that the injector is functional?

The PoE testers I mentioned earlier can detect the passive PoE but the indications can be a little difficult to fathom, particularly if plugged directly into the Tycon because of the swapped wiring. Best is to make a breakout RJ45 cable - just a short length of CAT5e connected to an RJ45 (with the standard wiring) and the other end bare wire ends. Then you do these checks with the RJ45 plugged in to the Tycon PoE out:

  1. Without power to the Tycon check the resistance between each color pair. It should be around 1 ohm or a little less. Anything over a couple of ohms indicates a bad connection or blown ethernet magnetics in the Tycon.
  2. Without power to the Tycon validate the magnetics by using standard RJ45 cables and plugging one end into a switch, or anything with LEDs to indicate a satisfactory link and the other into a computer or any device with an RJ45 port (including another switch). If you don't get a link the magnetics are probably blown. If you use devices on both ends which can do gigabit ethernet the test is much better; you should get a gigabit connection through the Tycon.
  3. Now go back to the breakout cable and very carefully separate all 8 wires then plug the power in to the Tycon. Take great care not to short the wires! You should see 48V across any two conductions from different ethernet channels when the power is connected. Channel 1 (1,2,3,6 IRC, but check) should be the +48V and channel two (the other four conductors) should be the -48V, so {1,2,3,6} to {7,8,4,5} should measure 48V for each combination (1,7 1,8, 1,4 1,5, etc, etc. It's a bit tedious to do the test but it is worth it.)

It is also possible to do a load test after (3) by using some load to verify that you can get 2A out of the setup, or to run a load tester as I think I described above.

The injector is passive. It doesn't use the power to do anything other than supply power on the PoE port so the channels can be tested without power. The power supply inside the Tycon is pretty basic so you can test it without any ethernet signals. This is why the above works.

Any thoughts on why my setup might be slowly frying the injector?

But is it fried? My own system drops out regularly but it is nothing to do with the PoE; what happens is that the dish gets an obstruction at the moment my router is trying to renew the DHCP lease from StarLink, this causes my router to permanently drop the connection until I pull out the RJ45 at the router and plug it back in again. It's a known problem (not necessarily StarLink's, since the DHCP dropout is pretty much unavoidable).

As for your setup it is certainly capable of frying the Tycon and maybe even the dish. The Tycon does not have any protection so connecting a hot PSU to it results in a significant, maybe massive, inrush of current until the dish converters are up to power. The short circuit protection in the boost converter may be enough to prevent damage, but who knows? If it were me I would use the 12->48 3A model; it should not be able to fry the Tycon unless something in your setup is continuously drawing 3A (or so) though the injector.

I can't see why the injector would end up damaged over time. I measured the current over a reasonably extended period (an hour or so) and saw peaks of 3.4A (as described above). Consequently if the injector is getting fried over time I would suspect that the boost converter is being erratic and sometimes delivering less than 48V. Because it is a DC boost converter the designer assumed that the 12V would be there all the time; it isn't designed to handle the 100-times-per-second dropout in the supply (input) voltage that happens with an AC converter. If your supply does drop out then it is possible that the Tycon will see repeated high inrush currents. You might test with a 12V battery (e.g. use a car jump starter, which should have enough capacity for close to 6 hours). If that works suspect a power supply issue.

It is very clear from everything that has been said that getting a DC-DC boost system to work is a lot more difficult than getting an AC-DC buck system to work. I have ideas about why this might be because of the differences between DC and AC PSUs but I'm just guessing. There must be an answer of course.

I have setup at least 100 of these both with ac-dc and dc-dc and have had 100% reliability. But I don't use tycons, I think they are junk compared to other alternatives.

And I never use 48v in most cases dishy end is getting undervolted with 48. I know the gen2 AP is 48 but it would be more reliable with slightly higher. I think 48 was done for cost savings not reliability.

@bghira
Copy link

bghira commented Apr 30, 2023

i had the same issue with the tycon. i tried a mccown but they're sensitive and i am pretty sure i sparked this one with a fat-fingered move where i accidentally brushed the shielded RJ45 connector up against something important - most likely an IEC. it was doing okay for a while after that, and even hit 190mbps but the next day it's limited to 10mbps.

currently i've given up on the idea of running it from purely 12v. it seems to work for some, but the heartache and expense and continually smelling burnt components just isn't sustainable or healthy. and i'm the type who usually refuses to give up until it works. i have a 1980 watt solar installation on my rooftop of my small house. was pretty frustrated with myself for shorting that McCown, as it took soo long to acquire.

@bghira
Copy link

bghira commented Apr 30, 2023

i will say the one thing i know is if you use the variable 1500w power supply from Amazon and tune it to 50v you'll have the best luck, as someone else had recommended. i followed that advice and that was the best. with the 388w or so, "heatsink resin style" transformer unit from Amazon that you're using, I just couldn't get any kind of reliable link from it, limiting to 1.5mbps or so. pretty abysmal.

@morehardware
Copy link

morehardware commented Apr 30, 2023 via email

@dougbrouwer
Copy link

Thanks Morehardware--Tycon makes a bunch of different PoE injectors. Is the cheap, POE-INJ-1000-WT that has been working for you, or are you using one of their more expensive models, and if so, which one?

Given experiences like yours, it's a mystery to me why swapping out the POE-INJ-1000-WT fixed my issue for a short time, and then failed, presumably, but not yet definitively, because of the injector again.

I've ordered one of the 1500W power supplies....

@bghira
Copy link

bghira commented Apr 30, 2023

well my Tycon was ruined by the Amazon cheap-ass PSU. when opening it up. the feet of the power module are separated from the PCB. they're not soldered very well and the heat pops them off.

@bghira
Copy link

bghira commented Apr 30, 2023

@dougbrouwer that's the one that'll work.

@dougbrouwer
Copy link

dougbrouwer commented Apr 30, 2023

"the feet of the power module are separated from the PCB."
Wow! That's pretty strange for a 130W rated device that carries a 3 year warranty.....
If Tycon doesn't replace mine under the warranty I'll open it up and report back.

@jbowler
Copy link

jbowler commented May 1, 2023

@dougbrouwer do you have any of the four connections to the buck converter connected (directly or indirectly) to ground? By "ground" I mean a structure of vehicle ground that part of the dish might be connected to? For example if the mast is grounded or the shielded cable from the dish. The Tycon shouldn't be connecting the shield to anything; it just passes the connection straight through, but if you have a surge suppressor inline or if you connect to a shielded port on the router that probably would.

@dougbrouwer
Copy link

any of the four connections to the buck converter connected (directly or indirectly) to ground?

Not any any way that's intentional. I'm simply using shielded RJ-45 connectors, and I'm being careful to connect the wire ground to the shield.

@dougbrouwer
Copy link

well my Tycon was ruined by the Amazon cheap-ass PSU. when opening it up.

@bghira What's the secret to opening up the Tycon injector? It's not obvious to me. Is it glued shut?

@jbowler
Copy link

jbowler commented May 2, 2023

any of the four connections to the buck converter connected (directly or indirectly) to ground?

Not any any way that's intentional. I'm simply using shielded RJ-45 connectors, and I'm being careful to connect the wire ground to the shield.

I'm not sure what you mean, but if you manage to ground the PoE then it is quite possible that you will blow it.

@bghira
Copy link

bghira commented May 4, 2023

@dougbrouwer a dremel

@Shangrila385
Copy link

Shangrila385 commented May 5, 2023 via email

@jbowler
Copy link

jbowler commented May 5, 2023

Can anyone explain in layman terms how to make a cable to use on starlink that is 300’ long and will work without losing signal strength? The dish is being put on the roof of a high rise bldg. And cable dropped down a shaft to lower floor.

That's actually very easy assuming there is a weatherproof location with power within 150' (or, cheaper, 75') of the dish location. That's pretty much a cert for a highrise. You put the StarLink router in the weatherproof enclosure, connect it to the building supply and connect it to the dish using the standard StarLink cables and the standard (USD25) Ethernet Dongle also from StarLink. Put the router into bypass mode, plug a standard 100m (300') ethernet cable into the dongle and the other end into the router on the lower floor.

There are lots of other ways, including the recently advertized stuff on Amazon that I quoted recently, but they are more expensive and, in most (but not all) cases, involve cutting cables. Easiest is to stick the router on the roof and do the drop from the dongle.

Anyone familiar with ethernet setup will know how to do this once they get their head round the weird StarLink router/dongle arrangement.

@Shangrila385
Copy link

Shangrila385 commented May 5, 2023 via email

@JustOneGuyHere
Copy link

I have a bunch of questions about Starlink cables. Not a regular here at all, and not a techie but also not a total idiot. I don't know if this is the right thread. I'm here because someone posted the link in a Starlink discussion. I don't want to start in on it for nothing.

@jbowler
Copy link

jbowler commented Jul 22, 2023

@JustOneGuyHere; it's a gist about getting a cable longer than the official StarLink 150ft cable for where the dish (the rectangular one, the v2 IRC) is more than 150ft from a power source. Since it's about power delivery the gist described how to build a new cable (by cutting the original one) with an additional power (PoE) injector. As a result it covers the wiring of the cable (standard ethernet, non-standard PoE). It applies to other cases where the cable is not that long but where it is desired to get rid of the white "tombstone" router. Things other than PoE requirements are incidental; there's some discussion about other things (e.g. I described one, apparently, common failure mode of the system) but it's slightly off-topic.

@JustOneGuyHere
Copy link

JustOneGuyHere commented Jul 22, 2023

@JustOneGuyHere; it's a gist about getting a cable longer than the official StarLink 150ft cable for where the dish (the rectangular one, the v2 IRC) is more than 150ft from a power source. Since it's about power delivery the gist described how to build a new cable (by cutting the original one) with an additional power (PoE) injector. As a result it covers the wiring of the cable (standard ethernet, non-standard PoE). It applies to other cases where the cable is not that long but where it is desired to get rid of the white "tombstone" router. Things other than PoE requirements are incidental; there's some discussion about other things (e.g. I described one, apparently, common failure mode of the system) but it's slightly off-topic.

I'm a civilian, and as such a little pathetic, but at least I admit it. I think I've learned that the Snow Melt feature winds up melting cables for many users (including us) and that there really isn't a realistic hack for it. I'm also standard user with a 50-foot cable and a dish mounted atop a 10-foot pole near the house. After replacing one cable with great difficulty, I've given up on Snow Melt and have turned it all the way off forever.

I'm told that even without the feature, what's in the dish uses 25-35 watts and generates some heat. Not enough to melt heavy snow (which we get sometimes) but I wonder if it'll be enough to keep ice from forming to begin with. Sorry if this is off-topic. If so, trust me, I'm not any kind of whackjob and will retreat right away. If you happen to be able to answer the ice question, I'd appreciate it. If you can't, I'll understand.

@jbowler
Copy link

jbowler commented Jul 23, 2023

I wonder if it'll be enough to keep ice from forming to begin with.

I don't know what failed in your/Starlink's kit, so bearing that in mind I'm just going to state the obvious.

Turn pre-heat on. It will cost you more in electricity but it might stop the snow-melt surge and that might be your problem. It's pretty much not possible to melt that cable, certainly not with the stock StarLink kit, even if the biggest surge I measured (which I don't quite believe) continued for a whole summer (when the cable resistance is highest).

In any case buy a power outlet tester and make absolutely sure the outlet the tombstone is plugged into is correctly grounded. If the outlet tester says anything weird call an electrician. Failing to ground the dish might result in static build up causing a static discharge that is sufficient to fry one of the four connectors. Failing to have adequate grounding in a house is much more dangerous of course.

Bottom line is that you have a failed piece of electrical equipment that indicates a major fault. I had the same problem and my conclusions are documented above, in a lot of detail; I try not to give conclusions, rather I document what I observe. This is github after all. I have my own conclusion which I don't think I've disclosed here but if you read what I said you can draw your own too.

@JustOneGuyHere
Copy link

JustOneGuyHere commented Jul 23, 2023

Bottom line is that you have a failed piece of electrical equipment that indicates a major fault. I had the same problem and my conclusions are documented above, in a lot of detail; I try not to give conclusions, rather I document what I observe. This is github after all. I have my own conclusion which I don't think I've disclosed here but if you read what I said you can draw your own too.

I will check the grounding, but strongly doubt this is the problem. The house was built 6 years ago, and there have been no electrical problems anywhere in the house. The speed first slowed down greatly, then disconnected altogether. After several weeks (typical Starlink -- terrible customer service) a new router and cable showed up.

The new router didn't work on the first cable, and now works fine on the second cable. The issue of fried cables turns out to be common. The four wire pairs are 24 gauge, which is fine until the thermostat in the dish tells the router to increase the amperage to the dish, which heats up the electronics inside the dish enough to melt snow on the dish face. Unfortunately, those very same 24 gauge wires tend to get fried by the extra current, especially in places where there are months of thermostat-dictated heating events, most of which (ironically) aren't even necessary because they were generated not by the presence of snow but only by cold temperatures.

At this point, I've decided to turn off Snow Melt, and only wonder whether 25-35 watts still going to the dish will generate enough heat to keep ice from forming. I know it won't keep snow from accumulating, so I will handle that issue by duct-taping a windshield scraper to an extender pole and wiping snow off. I'm told that the film on the dish is fragile enough to make it inadvisable to scrape ice off, but I wonder whether or not that 25-35 watts of power going to the dish will be enough to keep ice off, or whether I'll just have to turn on Snow Melt during storms and risk another cable failure.

You know, this wouldn't be happening if Musk had simply hired a graduate of Electrician 101 class at a community college to build a cable with 14 gauge wiring inside, but that's not rocket science so why bother? Better to irritate a lot of people and ship out new cables with the same design flaw. It's what happens when you're too cool for school. LOL

@JustOneGuyHere
Copy link

Tested for grounding. As expected, no problem. I have ordered a current meter. Based on some reading, I believe that the router operates at 110-120V and 0.25A without Snow Melt, and at 5A in Snow Melt mode. I don't have a meter yet, but will be getting one and testing it. Given how many cables are being fried by Snow Melt (I'm far from the only Starlink user who's reported the fault), I think the 24-26 AWG wiring in the cables is inadequate for the task.

I will check to be sure, but I am pretty convinced that Starlink under-engineered the cable wiring. As I see it, they went full Rocket Science on the satellites and the dishes, but ignored Electrician 101 when they designed the cables. The remaining question is for how long I can overload the wiring before I fry the wires inside the poorly designed cable.

@jbowler
Copy link

jbowler commented Jul 24, 2023

5A in Snow Melt mode. I don't have a meter yet, but will be getting one and testing it.

Please report back on that. I live in SW Oregon so getting the dish to go into snow melt was pretty much impossible.

@JustOneGuyHere
Copy link

JustOneGuyHere commented Jul 24, 2023 via email

@JustOneGuyHere
Copy link

JustOneGuyHere commented Jul 24, 2023 via email

@jbowler
Copy link

jbowler commented Jul 24, 2023

I also found a Kill-A-Watt meter, and in summertime (right now), it's drawing 122-123V and 0.35-0.4A with Snow Melt off or on automatic. In pre-heat, it draws 0.9A.

It sounds like you are measuring the current into the router. The power consumption of the router itself should be fairly constant. There will be an extra power draw from the dish, but you can work out the raw power required by the router simply by powering it on with the dish not connected (do not connect or disconnect the dish to a powered on router; this causes a very significant power surge to the dish which is the most likely candidate for frying one or more of the four connectors). See my earlier comment:

https://gist.github.com/darconeous/8c7899c4d2f849b881d6c43be55066ee?permalink_comment_id=4479847#gistcomment-4479847

I got my figures using the Tycon power injector described in this gist (not the router) so I was able to measure the current being sent down the cable. I also put a 2A fuse inline with the power delivery so that limits the short-term average current. The fuse never blew even though there is good reason to believe that there is a startup current/power surge which might be as much as 7A for a (small) fraction of a second (hence my comment about about not connecting a "hot" router to the dish). My detailed measurements start here:

https://gist.github.com/darconeous/8c7899c4d2f849b881d6c43be55066ee?permalink_comment_id=4463407#gistcomment-4463407

If you look at one of the posts that follow:

https://gist.github.com/darconeous/8c7899c4d2f849b881d6c43be55066ee?permalink_comment_id=4466245#gistcomment-4466245

You will see the graph of what actually happens during boot and after. This is using a Fluke 189 logging multimeter, pretty much the industry standard when it was bought for me it but I admit I've owned it for coming on 30 years and never got it recalibrated. The most telling figures might be the ones in the first comment: the dish apparently does consume close to 90W for short periods of time. 90W is the capacity of the router, some injectors can handle more but certainly not all of them.

I believe I left "pre-heat" on after my first post and you can see that the actual dish consumption hits a max of 1.6A but averages somewhere in the range 0.6 to 0.9A. This corresponds to a maximum power of 77W and an average over 5s in the range 29W-43W, consistent with the stated router capabilities.

It's almost impossible to guesstimate the actual power going to the router without a calibration of the router efficiency; the router takes 110-250V input and converts it to its own internal power requirements (5V or 3.3V, probably 5) and the dish requirements (48V). Efficiencies will be in the range 80-90% but the design might be dumb.

Nevertheless whatever excess current is delivered to the router when the dish is connected it is certainly an underestimate of the current going to the dish, because that current is at ~48V.

All the evidence so far is that the design of the dish produces significant surge currents under some circumstances. In particular there is evidence that there might be a very high initial surge if a charged, hot, PoE injector is connected to the dish. This is what I was doing with my Tycon setup and the tester I was using (which I don't trust) reported a 7A surge.

The surge should be no big problem for the components involved because they are mostly passive, maybe all; it's not clear how many diodes StarLink put into the circuit but it sounds like "none", that's one way of getting more power. My own tests prove that in my system the thing that fried was the connection (the connector) at the dish. It fried in a way that is consistent with an instantaneous surge. What caused that surge is impossible to say.

@JustOneGuyHere
Copy link

JustOneGuyHere commented Jul 25, 2023 via email

@jbowler
Copy link

jbowler commented Jul 25, 2023

This is consistent what what I've been reading (maybe here?) that the dish use 0.30A with Snow Melt completely off.

That's consistent with my numbers; it would correspond to about 0.6A going to the dish. A rule of thumb assuming 80-90% efficiency in the router is that the current flowing to the dish MIGHT be about twice the extra current to the router. Maybe "pre-heat" increases it slightly but I suspect that will be thermostatically controlled (this doesn't cost enough money for StarLink to cost-reduce it out of the dish). I measure in February, before the current heat but it was still well above 0C.

Our dish is always connected, so what does that mean? It's not as if I climb the ladder and monkey around with the connector.

Don't start up, power cycle, the router without the dish cable connected. I.e. don't plug the dish cable into the router when the router is connected to the power supply; this causes a surge because the router socket is already powered up and the dish is waiting for juice. It says this in the instructions; if you follow the instructions step by step the router is connected to the power after the router is connected to the dish. Of course you have to look at this like a lawyer or a computer programmer to see that as a clear instruction.

@JustOneGuyHere
Copy link

JustOneGuyHere commented Jul 25, 2023 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment