It's a problem with the macs, there's aprogram called Ejectify, but that doens't alwasy work for me.
There's probbaly other solutions as well (software)
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Thunderbolt 3 M.2 NVMe SSD Enclosures
- Thread starter oreggero
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
It's a problem with the macs, there's aprogram called Ejectify, but that doens't alwasy work for me.
There's probbaly other solutions as well (software)
I get this all the time; I ignore the warning and there appears to be no downside.
I don't know – it certainly hasn't always been like this. And it seems some enclosures/external drives don't have the problem. I never had this issue with the external enclosure for a 3.5" drive that I use. Hmm…
Also, if I want to back up the contents of the external drive to elsewhere I can't have it eject regularly.
There is a solution…
Quote:
Orico themselves says when I email them:
For your Mac and SSD, We recommend you buy our Thunderbolt chipset JHL7440 enclosure with widely compatibilities.
The reason for this is that Apple and Intel were the co-developers of Thunderbolt, and Intel still requires Certification of TB products.
If USB products are incompatible with Apple’s implementation of USB4 then I don’t think Apple is incentivised to troubleshoot except with manufacturers who have a developmental relationship with Apple - those who Apple regularly deal with.
Thunderbolt 3 devices like docks take over the control of their USB 3.2 Gen 2 or lesser channels and the incompatibilities are mostly gone.
Quote:
Orico themselves says when I email them:
For your Mac and SSD, We recommend you buy our Thunderbolt chipset JHL7440 enclosure with widely compatibilities.
The reason for this is that Apple and Intel were the co-developers of Thunderbolt, and Intel still requires Certification of TB products.
If USB products are incompatible with Apple’s implementation of USB4 then I don’t think Apple is incentivised to troubleshoot except with manufacturers who have a developmental relationship with Apple - those who Apple regularly deal with.
Thunderbolt 3 devices like docks take over the control of their USB 3.2 Gen 2 or lesser channels and the incompatibilities are mostly gone.
If that truly is the solution, then good.
I will first try with a different 40 Gbps capable USB-C cable and see if that makes a difference with the enclosure I already have.
iMac M3 8/256
TBU401E working great, no eject notices
Mac Studio M2 Ultra 64/1TB:
Asmedia 246x > Routinely gives eject notices (with no clear other downside)
TBU401E working great, no eject notices
Mac Studio M2 Ultra 64/1TB:
Asmedia 246x > Routinely gives eject notices (with no clear other downside)
Edit:
OK, the below is not true!
It took some more time but eventually now had disconnected when waking the MacBook from sleep with the enclosure connected. 🙁
Wanted to inform you all that using a new cable with my Orico ASM2464PD enclosure I've not had a single disconnect so far! I even let the case be attached overnight while the MacBook was sleeping and it was still there when I woke up the Mac. So it seem the cable can cause this problem.
I went from the cable that came with the enclosure (which also says 40 Gbps on it) to this:
www.belkin.com
The enclosure do get crazy warm when using it, though. Don't feel too comfortable about that.
OK, the below is not true!
It took some more time but eventually now had disconnected when waking the MacBook from sleep with the enclosure connected. 🙁
Wanted to inform you all that using a new cable with my Orico ASM2464PD enclosure I've not had a single disconnect so far! I even let the case be attached overnight while the MacBook was sleeping and it was still there when I woke up the Mac. So it seem the cable can cause this problem.
I went from the cable that came with the enclosure (which also says 40 Gbps on it) to this:
USB4 Cable - 2.6ft, 40Gbps 100W USB4 Type C Cable | Belkin | Belkin US
Access faster charging, HD video, and quick data transfer speeds of up to 40Gbps with our 100W USB4 Cable. Backward compatible with USB-C cables. Shop now.
The enclosure do get crazy warm when using it, though. Don't feel too comfortable about that.
Last edited:
Yeah the cables are an unexpected but common source of problems. Most people don’t consider replacing them, but after having several issues with supplied cables I think it’s best to keep at least one high quality one around to diagnose issues.
It' Not that because even some cheaper SSDs tend to sleep after some times not being used! Plus some these drives (especially brand model) tend to sleep. They seem to wake up when a Windows computer but on Mac/Linux machines the drive will not wake up to those Mac/Linux machines! Hence the 'disk not properly ejected' error on our Macs! Hence the error will continue unless you eject the drive before sleep every evening to this error!
I purchased the Jeyi 2464 40gbps NVME SSD with fan. I installed Samsung 980 pro and connected it to my iMac 2020 thunderbolt 3 port and tested it. The firmware was also updated to latest version. If I start my iMac with the enclosure connected the speed are 2940 read and 1650 write. If I plug in the enclosure when the iMac is running, after it has finished booting the speeds are 3085 read and 2960 write. I have tried different cables and connecting to a powered dock but the results are the same! Is there something I can do to change this or is it normal?
While the 2TB in my '23 Studio hovers around 33C, my TBU405 and Maiwo externals hover around 45C (which is about the same as the daily I recently experienced in the Mojave).
Don't know what your temps are, but I easily survived 40C+; so, don't worry too much about your metal-ceramic devices
Wanted to inform you all that using a new cable with my Orico ASM2464PD enclosure I've not had a single disconnect so far! I even let the case be attached overnight while the MacBook was sleeping and it was still there when I woke up the Mac. So it seem the cable can cause this problem.
I went from the cable that came with the enclosure (which also says 40 Gbps on it) to this:
USB4 Cable - 2.6ft, 40Gbps 100W USB4 Type C Cable | Belkin | Belkin US
Access faster charging, HD video, and quick data transfer speeds of up to 40Gbps with our 100W USB4 Cable. Backward compatible with USB-C cables. Shop now.
The enclosure do get crazy warm when using it, though. Don't feel too comfortable about that.
You need something like these pads to put on NVME device!
Thanks for the info! It does seem to not get as hot as before. Not sure why. Could the other cable that made it disconnect contribute to that? Maybe Spotlight was indexing it or something. Before it was close to me not being able to hold it in the hand for long. Now it’s not as bad. 40-45 C sounds like it could be for me too.While the 2TB in my '23 Studio hovers around 33C, my TBU405 and Maiwo externals hover around 45C (which is about the same as the daily I recently experienced in the Mojave).
Don't know what your temps are, but I easily survived 40C+; so, don't worry too much about your metal-ceramic devices
@fibercut
There was a similar pad included with the enclosure that I did put on the drive.
Welcome.
There's a lot of perspective out here. One person's awful commute is another's day off
Temperature analysis is just like blood pressure, and one reading does not define a trend.
The ASM2464PD chipset uses nearly twice the power at idle on macOS than it does on Windows. It's very obvious as the enclosure becomes too hot to touch with macOS, while with Windows it's only warm. I've posted about this on Reddit:
Top: HAGIBIS USB4 NVMe Enclosure
Bottom: MAIVO USB4 NVMe Enclosure - K1717
SSD: Samsung 990 Pro
Laptop: MBP 16" M1 Max - macOS 14.6.1:
Drive Temp at Idle: 55-65C
Surface Temp at Idle: 45-50C
Idle Power Usage: 5.5W
Active Power Usage: 7.5W
Laptop: Dell Precision 5560 - Windows 11:
Drive Temp at Idle: 42C
Surface Temp at Idle: 38C
Idle Power Usage: 3.2W
Active Power Usage: 7.5W
Top: HAGIBIS USB4 NVMe Enclosure
Bottom: MAIVO USB4 NVMe Enclosure - K1717
SSD: Samsung 990 Pro
Laptop: MBP 16" M1 Max - macOS 14.6.1:
Drive Temp at Idle: 55-65C
Surface Temp at Idle: 45-50C
Idle Power Usage: 5.5W
Active Power Usage: 7.5W
Laptop: Dell Precision 5560 - Windows 11:
Drive Temp at Idle: 42C
Surface Temp at Idle: 38C
Idle Power Usage: 3.2W
Active Power Usage: 7.5W
Are you on the latest firmware? I've been using ASM2464 enclosures for over a year and this was a common problem in earlier firmware. You can download the latest firmware from station-drivers and flash it, takes only a few minutes.
Yeah, that's a firmware bug with USB4 TB3 alt mode. I've experienced the same behavior on my iMac 2020 on every single firmware version released till date.
Last edited:
I suspect it's something to do with ASM2464PD firmware not properly implementing NVMe power states on the macOS side. The earlier firmwares used to be way worse. I tested the earliest firmware back in August 2023, and the idle power draw hovered around 7W.
Check the power draw of the empty enclosure connected to your Mac. It would most likely be ~3W, which would mean the drive itself consumes ~2W at idle due to poorly implemented NVMe power states on macOS.
Last edited:
USB4, NVMe, PCIe, and ASPM are all standards. Why would the chipset have issues on macOS but work fine on Windows when both should be using the same standards? Was it an earlier ASM2464PD firmware or an earlier build of macOS?
Earlier ASM2464PD firmware. I tested it on latest builds of macOS Monterey and Ventura at the time.
In theory it should be using the same standards, but the implementation can always require fine tuning. The earlier firmwares had a lot of glaring issues with macOS that were resolved in the subsequent releases.
I replaced the thermal pad in the Zike drive, which always runs hot, around 60°C idling. Have a Lexar NM790 4TB in there. I figured the original one shipped with the Zike drive perhaps wasn't thick enough. So I bought a stack of them. Added them until the case wouldn't close anymore and backed off by 1 pad. 48°C. Then when it was busy being written to, it went up to 58°C. Then, after being read for Time Machine, 56°C. Not a big difference. The M1 Max internal SSD is about 20°C cooler.
No amount of thermal pads are going to counter the chipset not idling. These enclosures don't have the surface area and mass to dissipate nearly 6W of heat at idle. It takes a good 30min to heat saturate the slab of aluminum.
Plug the Zike into a Windows 11 TB4 laptop and check the idle temps after 30min. You can actually hold the enclosure in your hand without risking burns.
I've visited this thread off and on, and neither do I.
How does that impact things if you use an external TB3 SSD as your startup disk?
Thank you for posting a temp. of a Mac's internal SSD vs. an external SSD.
The question that follows is what temperature range for each is appropriate, and at what temp.s should be get concerned?
The next question is what are the anticipated long-term consequences of an SSD running at 45 degrees C (113 F) instead of 33 C (91.4 F)?
Maybe so, if TB5 rolls out widely. But will TB5 becomes ubiquitous where Thunderbolt is used, the way TB3 and 4 did? Or will it be largely confined to higher end systems?
We're used to seeing different implementations of USB-C ports (e.g.: data and/or power delivery, transfer rates, Thunderbolt vs. not). Might we see that with Thunderbolt?
Are enough people using Thunderbolt (or USB-4) monitors to drive improvements? I'd like to see 5 or 6K 32" monitors at moderate (i.e.: Non-Apple) prices.
So now we've got a report of 60 degrees Calcius (140 F). You went to considerable DIY bother to get it down to 48 idling (118.4 F). Most people won't do that. How much long-term benefit is likely to be gained from what you did?
I sought an objective opinion on how hot 140 F is. From JohnsManville:
"Is 140 degrees hot to the touch?
ASTM C1055 (Standard Guide for Heated System Surface Conditions that Produce Contact Burn Injuries) recommends that pipe surface temperatures remain at or below 140°F. The reason for this is that the average person can touch a 140°F surface for up to five seconds without sustaining irreversible burn damage."
Why don't they?
I get it with the little portable external SSDs, but why is there not an external Thunderbolt 3 SSD designed for the desktop with enough mass to keep the thing cool?
If it stays cool in a Mac Studio, why can't it stay cool in an external enclosure?
Given Apple's severely high internal SSD upgrade prices, I imagine many users will ask whether it's better to buy an external TB3 SSD and make it the startup disk, saving several hundred, hoping that roughly halving the read/write speeds won't affect the user experience noticeably (speaking of which, what do you guys think about that?). But many people keep Macs a long time (I'm over 7-years old on an iMac now); over that time frame, what does an extra 12 - 27 degrees Celsius heat do?
P.S.: If I ever get to upgrade this iMac, the internal vs. external option will be on my mind.
The Zike enclosure was a first-gen thing with ASM2464PD. It has been very reliable, no disconnects. The cable is ridiculously short though. I have ordered a good quality 30cm cable. They also have a firmware update designed for Windows. I should probably install that and retest. But reliability is very important to me, so usually I don't update unless something is wrong.
Update:
I was wrong. Now today it had disconnected when I woke the MacBook from being asleep over night even with the new cable.
Will try the firmware update and see if that helps.
As I posed here the disconnecting stopped when I switched to another cable.
But interesting point about firmware. I'll see if I can give it a go. I hope I can trust Station Drivers.
Always a bit skeptical about downloading drivers from a third party. Seems pretty common, though. The latest AMD|Mediatek drivers I need for my Windows 11 desktop I built in the beginning of this year is offered via some individual hosting them on mega.nz. 🤷♂️
Do you know if I have to format the drive in the enclosure or can the firmware be updated independently?
Last edited: