Notifications

Active TB3 Cable overheating?  

  RSS

teltersat2
(@teltersat2)
Eminent Member
Joined: 6 months ago
 

Hey guys, I recently got a Razer Blade Stealth 2016 to tinker with and I wanted to try its TB3 capabilities. The testing got cut short as the actual TB3 link kept resetting (i.e. offline/online) every so often that made my current rig unusable. This would be the second computer to exhibit this behavior (1st one was a Lenovo T470s) , and also found out that the actual TB3 connector on the laptop side gets hot to the touch (not to the point of burning, but it is very hot indeed) - my MBP 2017 does not have any problems handling this. But now that I've expanded my laptop selection, I cannot help but think that it may be the cable itself. I have a Cable Matters Active 2.0m cable, and not sure if I should try another. I tried buying a CalDigit cable from Amazon, but that was a flop too, as it came in with a package that was ripped open and the cable itself was not working. Which cable would you recommend?

Current build: 2017 15" MacBook Pro + Mantiz Venus MZ-02 + Silverstone SX700-LPT (750W Peak) + Noctua NF-A14 + Gigabyte GeForce GTX 1070 G1 Gaming 8G (rev. 1.0)
Replaced parts: Sapphire Radeon RX 580 Nitro+; Sapphire Radeon RX Vega 64 (Reference model)


ReplyQuote
itsage
(@itsage)
Illustrious Member Admin
Joined: 3 years ago
 

@teltersat2 Can you see the Thunderbolt 3 Controller ID for your 2016 Razer Blade Stealth? I typically set up a new host with the .5m cable to make sure the components don’t cause issues. I have used longer than .5m Thunderbolt 3 cables from AKiTiO, ASUS, and Razer. They all work reliably. The connected laptop’s Thunderbolt 3 port area can get warm from time to time like you said.

 

external graphics card builds
best laptops for external GPU
eGPU enclosure buyer's guide


ReplyQuote
teltersat2
(@teltersat2)
Eminent Member
Joined: 6 months ago
 

Will do when I get home - I did manage to get enough time (10 minutes) to set up everything and getting all the devices recognized. After that, the laptop started connecting/disconnecting and that's when I stopped. I do have a passive 0.8m cable, but since my eGPU is routed underneath my desk I'm going to have to get creative for a bit to test it.

Current build: 2017 15" MacBook Pro + Mantiz Venus MZ-02 + Silverstone SX700-LPT (750W Peak) + Noctua NF-A14 + Gigabyte GeForce GTX 1070 G1 Gaming 8G (rev. 1.0)
Replaced parts: Sapphire Radeon RX 580 Nitro+; Sapphire Radeon RX Vega 64 (Reference model)


ReplyQuote
wimpzilla
(@wimpzilla)
Honorable Member
Joined: 3 years ago
 

I would avoid using this cable any further, it could be faulty and kill the TB controllers when it dies.
If it heats like this, i would simply suppose that some wire or active components begin to short to ground.
Leading to a massive leak of current causing the heat you feel on the plug.

2012 13-inch Dell Latitude E6320 + R9 [email protected] (EXP GDC 8.4) + Win10
E=Mc²


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 1 year ago
 
Posted by: @wimpzilla

I would avoid using this cable any further, it could be faulty and kill the TB controllers when it dies.
If it heats like this, i would simply suppose that some wire or active components begin to short to ground.
Leading to a massive leak of current causing the heat you feel on the plug.

I thought the concept of "short to ground" only applies for high voltage and AC voltage. Neither one is in my TB3 cables.

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
teltersat2
(@teltersat2)
Eminent Member
Joined: 6 months ago
 

@itsage

So I have the following controllers in my laptops:

  • Thinkpad T470s: Thunderbolt controller - 15BF
  • Razer Blade Stealth: Thunderbolt Controller - 15D9

Is this the right ID?

Was also thinking that it could be the power delivery portion of the cable: 100W vs 60W rated cables. Not sure if the Thinkpad and the Blade can withstand all of that wattage on their PD circuit. Also, not sure if this has nothing to do with it either 🙂

Current build: 2017 15" MacBook Pro + Mantiz Venus MZ-02 + Silverstone SX700-LPT (750W Peak) + Noctua NF-A14 + Gigabyte GeForce GTX 1070 G1 Gaming 8G (rev. 1.0)
Replaced parts: Sapphire Radeon RX 580 Nitro+; Sapphire Radeon RX Vega 64 (Reference model)


ReplyQuote
joevt
(@joevt)
Noble Member
Joined: 2 years ago
 

@teltersat2 The cables have a chip that tell the controller how much power they can handle so it shouldn't be a problem. All the pieces negotiate with each other so that only the max power allowed by all the pieces is delivered.

I would think a short would cause immediate failure and no power would be delivered. Maybe the resetting you experience is the failure. In that case, try a different cable. If a new cable works, then it's time to throw out the old cable.

Mac mini (2018), Mac Pro (Early 2008), MacBook Pro (Retina, 15-inch, Mid 2015), GA-Z170X-Gaming 7, Sapphire Pulse Radeon RX 580 8GB GDDR5, Sonnet Echo Express III-D, Trebleet Thunderbolt 3 to NVMe M.2 case


ReplyQuote
teltersat2
(@teltersat2)
Eminent Member
Joined: 6 months ago
 

@joevt
That's what I thought, but I thought the Power Delivery control would be in the device rather than in the cable. TIL!
In any case I have decided to order another cable, which hopefully will arrive intact and in a working condition!

Current build: 2017 15" MacBook Pro + Mantiz Venus MZ-02 + Silverstone SX700-LPT (750W Peak) + Noctua NF-A14 + Gigabyte GeForce GTX 1070 G1 Gaming 8G (rev. 1.0)
Replaced parts: Sapphire Radeon RX 580 Nitro+; Sapphire Radeon RX Vega 64 (Reference model)


ReplyQuote
joevt
(@joevt)
Noble Member
Joined: 2 years ago
 

@teltersat2

The cable also has information about whether it's for Thunderbolt (20 Gbps) or USB 3.1 gen 2 (10 Gbps) or USB 3.0 (5 Gbps). I suppose Virtual Link would have a unique identifier as well in its cables (does anything use Virtual Link or is it dead?)

Mac mini (2018), Mac Pro (Early 2008), MacBook Pro (Retina, 15-inch, Mid 2015), GA-Z170X-Gaming 7, Sapphire Pulse Radeon RX 580 8GB GDDR5, Sonnet Echo Express III-D, Trebleet Thunderbolt 3 to NVMe M.2 case


ReplyQuote
wimpzilla
(@wimpzilla)
Honorable Member
Joined: 3 years ago
 

@OliverB

At DC levels one define the resistance as an object that oppose current flow.
At AC levels one define the impedance as an object that oppose current flow.
A short to ground define the shortest path in a circuit with the lowest resistance, which mean the highest current flow.

At DC levels the current flow in one direction in a conductor, imagine a pipe already full of water where you would push more water in, into only one direction.
At AC levels the current flow in one direction, then come back in the other direction, imagine a pipe already full of water where you would push the water from one side and then in other side.
Note that at DC levels one need to push the water out of the tube constantly, at AC level one just bounce back and forth the same water, so AC have the best efficiency!

So a short to ground at DC levels is not the same at AC levels.
The best example of that is an antenna, it is like a grounded loop of wire shorted at DC levels but instead produce a signal when appropriate modulation and frequency is applied.
Each fundamental characteristic of circuits like, voltage, current, resistance, capacitance, inductance do not behave the same under AC/DC load.
AC define the variation of current flow into the conductor, the current flow variation oscillate with a frequency.
Nevertheless one would still get an AC component at DC level, like the tiny variations of peak voltage/current that happen on top of your 12v DC one direction current flow.
Nowadays computers and communications data lines follow much more AC rules than DC rules, i think it answers your question.

It is a quite vast field, just remember that everything have a frequency of oscillation, it's not because you don't feel it that it don't vibrate.
Also always remember that a defined oscillation frequency can always be represented with the sum of multiples others different frequency oscillations.

EDIT:
@OliverB
The ground "plane" could have its own resistance because it's a big sheet of metal.
But what drive the current flow is the difference of voltage potential, one define the ground as the lowest voltage potential point of the circuit.
It could happen to be 0v or 100v or even -500v it doesn't matter, what matter is how much resistance there is between the voltage potential difference.
Simple test, get a 18600 battery, get a simple wire with almost no resistance and then short the battery, the wire will melt with almost 20A@3.6v passing through it.
Tho the battery have a metal shell that provide the ground plane with a define resistance, it didn't help to not burn your wire or make the battery die/explode.

This post was modified 5 months ago

2012 13-inch Dell Latitude E6320 + R9 [email protected] (EXP GDC 8.4) + Win10
E=Mc²


itsage liked
ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 1 year ago
 

@wimpzilla
Ground resistance is way too high for 5V. There will be no measurable current.

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
teltersat2
(@teltersat2)
Eminent Member
Joined: 6 months ago
 

Just an update: I just finished testing a Razer TB3 cable because I just thought it would be best to NOT go through Amazon, and I found out that given appropriate cooling conditions, the Razer cable would work fine whereas the Cable Matters one would not:

  • Lenovo T470s: Works without any hiccups on top of a mouse mat
  • Razer Blade Stealth 2016: Disconnects on top of a mouse mat, but works fine in a metal stand. This makes me believe that is both the cooling portion of the laptop + the cable. Laptop will go in a metal stand so no heating problems
  • MacBook Pro 2017: Works without any hiccups

Glad to have this resolved!

Current build: 2017 15" MacBook Pro + Mantiz Venus MZ-02 + Silverstone SX700-LPT (750W Peak) + Noctua NF-A14 + Gigabyte GeForce GTX 1070 G1 Gaming 8G (rev. 1.0)
Replaced parts: Sapphire Radeon RX 580 Nitro+; Sapphire Radeon RX Vega 64 (Reference model)


ReplyQuote
wimpzilla
(@wimpzilla)
Honorable Member
Joined: 3 years ago
 

Well, using a large mouse pad instead of directly the solid desk surface is not the best way to help any king of electronics to radiate heat.

Occasionally breathing a bit of air here and there.   😎 😉 

This post was modified 5 months ago

2012 13-inch Dell Latitude E6320 + R9 [email protected] (EXP GDC 8.4) + Win10
E=Mc²


ReplyQuote
teltersat2
(@teltersat2)
Eminent Member
Joined: 6 months ago
 
Posted by: @wimpzilla

Well, using a large mouse pad instead of directly the solid desk surface is not the best way to help any king of electronics to radiate heat.

Occasionally breathing a bit of air here and there.   😎 😉 

Indeed! But I did not think it would cause an issue within 1 minute 😛 Thankfully I did have a cooling pad at hand!

Current build: 2017 15" MacBook Pro + Mantiz Venus MZ-02 + Silverstone SX700-LPT (750W Peak) + Noctua NF-A14 + Gigabyte GeForce GTX 1070 G1 Gaming 8G (rev. 1.0)
Replaced parts: Sapphire Radeon RX 580 Nitro+; Sapphire Radeon RX Vega 64 (Reference model)


ReplyQuote
wimpzilla
(@wimpzilla)
Honorable Member
Joined: 3 years ago
 

Glad you sorted out.

Designing reliable products is not the main focus point nowadays, unfortunately.

Never expect anything within the tech industry, even what seems obvious, as common sense.  😉 😎 

 

 

 

This post was modified 5 months ago

2012 13-inch Dell Latitude E6320 + R9 [email protected] (EXP GDC 8.4) + Win10
E=Mc²


ReplyQuote