MacInTouch Amazon link...

USB, FireWire & Thunderbolt

Channels
Troubleshooting, Products
I have always experienced the random disconnects on my 2018 Mac Mini. With my recently purchased 2019 15" MacBook Pro, I'm experiencing those same exact disconnects. Regularly. (On my late 2017 5K iMac, curiously, I do not.)
According to the tech support rep who responded, the 2019 MacBook Pro has not been "certified" to work with the Envoy Pro. He also acknowledged the disconnect issue with the 2018 Mac Mini. What is really weird in his response is that he suggested turning off WiFi to see if that solves the issue. What?
See my previous post-7654 on the topic...
... After reading your earlier post, and as I stated, it was time to do a little experimentation.

Well, after a weekend of said experimentation I can report back that my iPhone 11 Pro Max was the culprit with the OWC Thunderbolt 3 drive. Using both the Mac Mini, which spontaneously disconnected regularly, and my MacBook Pro, I experienced not a single disconnect of the external drive after I placed my iPhone in a different area of my office. Not a single one! Once I placed the iPhone next to either the Mini or MacBook Pro, the drive spontaneously disconnected.

Which begs the question: why? Improper shielding of the Envoy Pro X? Sunspots? Lunar activity? :)
 


Ric Ford

MacInTouch
Which begs the question: why? Improper shielding of the Envoy Pro X? Sunspots? Lunar activity? :)
It would be interesting to try different cables to see if that might be a factor. But moving 40 Gbits per second is a tricky business electrically, as is cellphone and WiFi radio. It would also be interesting to try different iPhone modes - Airplane mode, WiFi on/off, cellular data on/off, Bluetooth on/off.

I've long wanted to do some serious Thunderbolt data integrity testing - before putting Thunderbolt into production - but haven't been able to get the the time to do it (and have just given up a 2018 MacBook Pro, though I still have access to a 2017 iMac with Thunderbolt 3).

I did encounter the problem once while briefly testing a Dell Inspiron laptop with a Samsung X5.
 


It would be interesting to try different cables to see if that might be a factor. But moving 40 Gbits per second is a tricky business electrically, as is cellphone and WiFi radio. It would also be interesting to try different iPhone modes - Airplane mode, WiFi on/off, cellular data on/off, Bluetooth on/off.
Unfortunately, Ric, the Envoy Pro has a non-detachable cable, i.e. it's permanently affixed.

Before I purchased the iPhone 11 Pro Max, I had an iPhone XS Max which I used to keep on my desk next to my other devices and had those same disconnects.

I might experiment further as you wrote above, time permitting.
 


Unfortunately, Ric, the Envoy Pro has a non-detachable cable, i.e. it's permanently affixed.
Maybe try an experiment surrounding the cable with a sleeve of household aluminum foil. I've seen this done in laboratory settings. I'm no expert, so no promises or explanations.
 





Interesting. Even more interesting because I keep my iPhone 11 Pro Max right next to the Envoy Pro. Time for some experimentation. Thanks, Ric.
I have my phone in a Qi charging stand and also a 2016 MacBook computer with Wi-Fi close to a few Thunderbolt 3 cables to monitors and docks. So far no issue with Glyph drive, Akitio RAID3 dock and drive, and LG 34wk95 monitor, along with Apple 27-inch Thunderbolt 2 monitor, a USB 3 hub and whole bunch of cables to boards that I debug.
 


Ric Ford

MacInTouch
I have my phone in a Qi charging stand and also a 2016 MacBook computer with Wi-Fi close to a few Thunderbolt 3 cables to monitors and docks. So far no issue with Glyph drive, Akitio RAID3 dock and drive, and LG 34wk95 monitor, along with Apple 27-inch Thunderbolt 2 monitor, a USB 3 hub and whole bunch of cables to boards that I debug.
The phrase "leading a charmed life" comes to mind....
 


The phrase "leading a charmed life" comes to mind....
This seems to be pointing to problems of electromagnetic compatibility, which can be resolved, but not necessarily easily. If your equipment doesn't have captive cables, your best bet may be to get "high performance", hence high-priced, cables.

You can also check the power and signal grounding system associated with your overall setup. You could try covering your desktop with a very well grounded copper or aluminum sheet that is insulated from you, in case of mishap. Remember that at these high data rates, any effective grounding is more like a wide strap or braid than a wire.

Generally speaking though, there is very likely a vendor quality problem that should have been revealed in their pre-production quality testing. Portions of the problem may be shared by several vendors.
 


Just got a MacBook Pro 16" with Catalina 10.15.2. When connected to an Akitio Thunderbolt 3 Dock with a dual drive in RAID0, it will always crash during sleep when the sleep is for several minutes. The crash occurs even when the drives are ejected before sleeping. I have unchecked put hard disks to sleep when possible and enabled power nap in the Energy Saver preferences, but it still crashes.

This issue did not occur on a 2016 MacBook Pro 15" with Mojave (10.14.6). OWC states this is a problem with Catalina that has not been resolved. To prevent crashes, they recommend to disconnect the dock before sleep.
This is not a new problem. I've been having a similar issue for years with an OWC Thunderbolt 2 Mini RAID with 4 SSDs. I have permanently disabled sleep on my 2014 iMac 5K (running Mojave), which minimizes the problem. But it still occasionally occurs when the computer is unattended (not sleeping). I come back to a "your computer has restarted due to a problem..." message. This happens every few days (3 - 5). The system log always points to the Thunderbolt Mini box. I also have a full size OWC Thunderbolt 2 RAID with four spinning drives and never had a problem with it. I just shut my system down when not used to try and minimize the problem. It's a pain.

Every once in a while the crash corrupts something on the RAID Mini. Usually, SoftRAID can fix it, but two months ago I needed DiskWarrior to rebuild the RAID to get it working again (a solution which I found in the SoftRAID forums).
 


The phrase "leading a charmed life" comes to mind....
I did find that I am having bluetooth interference issues with my Apple Magic Trackpad when I use 2-meter Thunderbolt 3 cables.... The trackpad has erratic jumps and starts, until I replace the 2m cables with the 6-inch Thunderbolt cable. Too bad those are too short, so I am using a USB cable connection to the trackpad...
 


The 22Gbps limitation is a limitation of the Thunderbolt controllers so far, not Thunderbolt protocol. There has only been one implementer company so far, [not] multiple implementers with perhaps differing design choices....
As far as I can tell (because there's no public specification published and no official statements), the 22 Gbps PCIe limitation on Thunderbolt 3 is entirely a protocol specification limit. The most pertinent technical details are contained in the Thunderbolt 3 Technology Brief, which has been referenced several times on this forum. Right off the bat, the brief confirms that Thunderbolt 3 attaches to only 4x PCIe 3.0 lanes, so that's a maximum of 31.5 Gbps that could be allocated to a PCIe connection through Thunderbolt. But that doesn't explain the difference between 31.5 Gbps and the 22 Gbps data limit that Thunderbolt 3 imposes. After looking into this 3 or 4 times now, I finally found this unsourced comment that claims the missing 10 Gbps is reserved for uninterrupted USB operation on the 2nd port. I don't know if that's true or not, but the 22Gbps speed limit is real, as has been borne out in all external Thunderbolt 3 storage benchmarks.
USB 4.0 adopts the base Thunderbolt networking protocol. That is a huge assumption that it is going to be better than the other Thunderbolt controller implementations on the first-generation implementations of USB 4.
USB 4 is going to try to route (encode and tunnel) three protocols, not just the two of Thunderbolt up to now. That isn't going to make low-latency tolerance for DisplayPort traffic get any better than what the Thunderbolt controllers have had to do.
So, the USB4 specification is actually available for download here. It's very dense and technical, and it's kind of crazy that I had to read this document to get the info we're looking for, instead of finding some accurate summary on the web somewhere. In any case, Section 2 gives a nice architectural overview. There is a "USB4 Gen 3x2" operating mode defined which gives the full 40 Gbps performance to the USB4 protocol (2.1.1.4.1). DisplayPort is mandatory, but PCIe is optional (2.1.1.5). Thunderbolt 3 backwards-compatibility is optional (2.1.5), so when it's operating in USB4 mode only, it will not be hampered by the Thunderbolt protocol restrictions I outlined above.

USB4 uses 128b/132b data encoding (4.3.2.1), the same as USB 3.1 Gen 2, whereas Thunderbolt 3 uses 8b/10b encoding, so USB4 only has a 3% protocol overhead vs. Thunderbolt's 20% overhead.

In scanning through the spec, I was unable to find a clear indication of how USB4 handles bandwidth allocation and prioritization, but everything I've read indicates that it's much more flexible than Thunderbolt. I think that when a USB4 Gen 3x2 external drive is connected to a USB4 Gen 3x2 host, that a data-only connection (with no DisplayPort or PCIe traffic) will be able to achieve near the 38.8 Gbps effective transfer rate. That means a PCIe 3.0 NVMe SSD in a USB4 Gen 3x2 enclosure would be able to achieve its full speed compared to a Thunderbolt 3 drive (probably about a 25% speed increase). And a PCIe 4.0 NVMe SSD in a USB4 enclosure would probably be almost 75% faster than a Thunderbolt 3 drive.

It might also be faster/cheaper for someone to create a USB4 SSD enclosure that attaches directly to the PCIe tunneling of USB4, rather than going through a PCIe-to-USB4 bridge inside the enclosure. But then they'd lose backwards-compatibility with USB 3.x and 2.0 hosts. And, of course, PCIe support is optional on USB4 hosts.
 



Of course, Thunderbolt 3 devices fail to work with USB-only ports, despite sharing the same Type-C connector, so does this change with USB 4?
No, Thunderbolt 3 backwards-compatibility is optional in USB4. So, for a Thunderbolt 3 device to work, it'll have to be connected to a Thunderbolt 3 host or a USB4 host that includes Thunderbolt 3 backwards-compatibility.

Since PCIe tunneling is optional in USB4, I suspect that any USB4 host that doesn't support PCIe tunneling will also not support Thunderbolt 3. I imagine that once Apple adopts USB4, all Mac USB4 ports will support PCIe tunneling and Thunderbolt 3 backwards-compatibility.
 


Ric Ford

MacInTouch
I've long known about Thunderbolt having various integrity/bug issues, but here's an interesting new example:
Tom's Guide said:
Nearly all ThinkPads have defective USB-C ports: What to do now

... ThinkPad laptops from 2017 to 2019 reportedly contain defective USB-C ports. First reported by Notebookcheck, the problems are widespread enough for Lenovo to post a support page titled "Critical Intel Thunderbolt Software and Firmware Updates."

Based on that document, the USB-C ports in more than a dozen ThinkPad laptops are experiencing serious issues. Lenovo posted a list of symptoms customers could encounter if they own a defective device:

Lenovo ThinkPad USB-C problems
  • USB-C port not working
  • Intel Thunderbolt controller not visible in the OS/Device Manager
  • USB-C or Thunderbolt docking stations not visible or having connectivity problems
  • HDMI output not available
  • System battery not charging with a USB-C power adapter connected to the USB-C port
  • Intel Thunderbolt pop-up error message
  • Intel Thunderbolt safe mode error message
  • BIOS Thunderbolt communication error or hang during POST
 



I bought one of these to replace an OWC 10-port dock. The problems I had with the OWC dock were slow charging, because it's only 60 watts, and the ethernet would not work unless I power cycled the dock with the MacBook Pro connected.

The ethernet on the CalDigit does not have this problem, and the charger supports 85 watts. However, sometimes I will plug it in and the MacBook Pro recognizes everything connected to the dock, but the battery says that it is not charging, and will in fact run down. Sometimes I can fix this by plugging into a different port on the Macbook Pro.

There is a firmware update for the CalDigit, and maybe that will help; however, it requires Catalina, which I cannot run (I still need some 32-bit apps, such as QuickTime). I am going to try to install Catalina on an external drive, boot from that, and update the firmware. What a PITA.
 


Ric Ford

MacInTouch
The ethernet on the CalDigit does not have this problem, and the charger supports 85 watts. However, sometimes I will plug it in and the MacBook Pro recognizes everything connected to the dock, but the battery says that it is not charging, and will in fact run down. Sometimes I can fix this by plugging into a different port on the Macbook Pro.
Note (per the user manual) that only the Thunderbolt 3 "Host" port provides power; the other Thunderbolt 3 port does not. (Also, only one non-Thunderbolt USB-C port is 10Gbps, and those don't provide laptop charging power.)
CalDigit said:
The TS3 Plus features two Thunderbolt 3 Type-C ports that can transfer at speeds up to 40Gb/s. One port is reserved for the connection to your host computer and the second can be used to connect extra Thunderbolt 3 devices. Thunderbolt 3 Type-C supports video and data. The Thunderbolt host port also provides charging up to 87W for your computer.

The TS3 Plus features two USB 3.1 Type-C ports (1 x 5Gb/s, 1 x 10Gb/s). These ports only support data and 1.5Amp (7.5W) of power for charging. Video capabilities are not supported.
 


The ethernet on the CalDigit does not have this problem, and the charger supports 85 watts. However, sometimes I will plug it in and the MacBook Pro recognizes everything connected to the dock, but the battery says that it is not charging, and will in fact run down. Sometimes I can fix this by plugging into a different port on the Macbook Pro.
I have three CalDigit TS3 Plus docks – one at home, one at work and one that is my wife's.

You plug your Thunderbolt 3 cable into the port on the back marked "[computer]" and it charges. Never has failed me. I leave the Thunderbolt 3 cable that came with the dock plugged in and unplug the MacBook Pro (2017 15" MacBook Pro with touchbar and macOS 10.12.6 Sierra) when I leave home and do the same at work. The cable never gets unplugged form the dock, just from the computer.

My only problem is that sometimes when coming home from work and plugging in, the MacBook Pro doesn't always wake up. The display is an LG 27UD68-P 27-Inch 4K UHD. I don't know if it's the display or something else. At work I have an LG 32UD59-B 32-Inch 4K UHD LED, and the MacBook Pro always wakes up when I get there and plug in.

Anyway, charging has never been a problem for me with this dock.
 


I have three CalDigit TS3 Plus docks – one at home, one at work and one that is my wife's....
I have one of these. The first one I received was basically DOA (it supplied power and the video signal passed through to the other Thunderbolt 3 port, but all the other ports didn't work). After a bit of a go-round with CalDigit support, I got a replacement that mostly works. Now all the ports work except that when I plug the 'computer' Thunderbolt 3 cable into my laptop, the video signal out of the other Thunderbolt 3 port doesn't turn on unless I remove and reinsert the monitor's cable. I haven't yet initiated a warranty replacement, but I will soon. Despite the "No Signal" issue with my external monitor, I do like it. I [had plugged] something into all 4 of the ports on my MacBook Pro (two monitors, ethernet adaptor, and power) – now it's almost a single cable to start my day.
 


Ric Ford

MacInTouch
The first one I received was basically DOA (it supplied power and the video signal passed through to the other Thunderbolt 3 port, but all the other ports didn't work). After a bit of a go-round with CalDigit support, I got a replacement that mostly works.
That sounds like it could have been a cable issue – for example, using a USB-C cable instead of a Thunderbolt 3 cable.
Now all the ports work except that when I plug the 'computer' Thunderbolt 3 cable into my laptop, the video signal out of the other Thunderbolt 3 port doesn't turn on unless I remove and reinsert the monitor's cable.
Given all the problems with Macs and external monitors, that may not have anything to do with the Thunderbolt 3 dock. Have you tried other displays or other ports (e.g. the dock's DisplayPort output)? Also, you might want to check the display's on-board settings (e.g. for sleep). What is the make and model of the display?
 


Note (per the user manual) that only the Thunderbolt 3 "Host" port provides power; the other Thunderbolt 3 port does not. (Also, only one non-Thunderbolt USB-C port is 10Gbps, and those don't provide laptop charging power.)
Turns out it was the cable I was using. I had previously purchased a 6-foot-long Thunderbolt 3 OWC cable (to use with the old OWC Thunderbolt 3 dock, and it worked fine). Apparently, in order to be rated for more than 60 watts over a certain length, you need an active cable. Using the short cable that Caldigit supplied, it works fine.
 


The first one I received was basically DOA (it supplied power and the video signal passed through to the other Thunderbolt 3 port, but all the other ports didn't work). After a bit of a go-round with CalDigit support, I got a replacement that mostly works.
That sounds like it could have been a cable issue – for example, using a USB-C cable instead of a Thunderbolt 3 cable.
Now all the ports work except that when I plug the 'computer' Thunderbolt 3 cable into my laptop, the video signal out of the other Thunderbolt 3 port doesn't turn on unless I remove and reinsert the monitor's cable.
Given all the problems with Macs and external monitors, that may not have anything to do with the Thunderbolt 3 dock. Have you tried other displays or other ports (e.g. the dock's DisplayPort output)? Also, you might want to check the display's on-board settings (e.g. for sleep). What is the make and model of the display?
I have two identical 4K LG USB-C/DIsplayPort monitors that were working fine, each with the cable with which it came plugged into a port on my MacBook Pro. Now I have one of them plugged into the TS3+ dock with a DisplayPort cable, and it works great. I have a couple of USB3 drives plugged into the dock, and they work great.

I have tried plugging the other monitor into the non-"computer" Thunderbolt 3 port with each of the monitor cables and a separately obtained Thunderbolt 3 cable. All three cables are able to drive the monitor, but not when the MacBook Pro has been asleep or been disconnected from the dock. Then I have to pull out and then reinsert the cable into the Thunderbolt 3 port on the dock.

The USB-C port on the dock does not supply a video signal (I asked CalDigit).
 



Amazon disclaimer:
As an Amazon Associate I earn from qualifying purchases.

Latest posts