[SOLVED] New SSD Temp going critical

[agalpha]

Hi
I just bought a brand new ADATA SSD 120GB 650SU model. I put the SSD in and cloned my current drive to it. Afterwards I ran CrystalDiskInfo, HD Sentinal and ADATA's own SSD tool.

CrystalDiskInfo - Temp: 55-57c(red), Health: Good.

HD Sentinal - Temp: 56c-57c(red), Health: 100%, Performance: 100%

ADATA SSD Tool - Temp 56c-57c, Remaining Estimated Life: 35%, Data Written 18886TB

All utilities show Power on count as of writing this as 9 and total hours as 1 hour (like I said, its a new SSD)

Crystal Disk Mark benchmark is perfectly fine and even better than my previous LiteOn SSD that I was using.

So why are the Disk Utilities showing critical temps and ADATA's own tool showing 18800TB written and 35% remaining life. EDIT: In another two hours the total Data Written to Disk is now showing 20840TB

Other drives in the system are running below 43c and SSD is infront of the front fan and in a cool place away from everything.

These temps are at idle with no file transfers or no heavy I/O programs running.

Screenshots

Should I be worried or is it just some bug in the firmware causing wrong reads of these values?

 

[jason201]

I'd say it's most likely a firmware bug. Buf if you wanna know for sure, you could use a multimeter (which has temperature measurement), or place your hand on it and feel it out.

 

[lynx1021]

My A-Data SU650 960GB show 41C and it is just hanging, not even mounted.Have you run diagnostics? I have had mine for 2 months and my estimated life is 99% from smart details. That estimated life on the tool box is back-wards, its showing how much you have used of the 100% life.

 

[agalpha]

I'd say it's most likely a firmware bug. Buf if you wanna know for sure, you could use a multimeter (which has temperature measurement), or place your hand on it and feel it out.

The drive is not hot to the touch. The other drives with lower temp readings are warmer when i touch them. The SSD is just normal

 

[agalpha]

My A-Data SU650 960GB show 41C and it is just hanging, not even mounted.Have you run diagnostics? I have had mine for 2 months and my estimated life is 99% from smart details. That estimated life on the tool box is back-wards, its showing how much you have used of the 100% life.

Can you show me a a screenshot of the Temps so I can at least show it to the retailer and try to get a replacement from him. If I can convince him of a fault he will simply replace the drive and keep the replacement he gets from the company. I am actually in Pakistan and getting replacements is a hassle much less a refund.

 

[fzabkar]

Can you show us the rest of CrystalDiskInfo's SMART attributes?

 

[lynx1021]

how do you post a picture here? Figured it out

https://imgur.com/a/85LlbOY

View: https://imgur.com/a/85LlbOY

 

[agalpha]

Can you show us the rest of CrystalDiskInfo's SMART attributes?

 

[agalpha]

I got a response from ADATA customer support regarding the temps and the wrong reading for estimated remaining life. Here is the response from them:

This is regarding the wrong estimated life.

This is regarding the high temps. They think it's normal and the drive is well inside the operating temperature.

 

[seanwebster]

It could be doing some background activity, which can cause it to heat up a bit. It will heat when transferring data, when OS writes logs/updates/edits files, or when garbage collection and wear-leveling kick in.

 

[DMAN999]

That may be normal for your SSD but IMO it's still too warm if that is the idle temp.
My standard MX500's (I have 3) all idle at around 30-32C and max out around 53C.
My m.2 SATA III MX500 idles at about 40 and maxes out in the mid-high 50's.
My OS SSD (Samsung 850 500 GB) idles at 29C and I have never seen it above 50C.
But I also have Excellent airflow in my case (2 front intake fans, a side intake fans and a rear and top exhaust fan).
And my MX500 in a USB 3 enclosure idles at 35C and maxes out around 60C under heavy use.

 

[lynx1021]

I got a response from ADATA customer support regarding the temps and the wrong reading for estimated remaining life. Here is the response from them:

This is regarding the wrong estimated life.

This is regarding the high temps. They think it's normal and the drive is well inside the operating temperature.

The A-Data tool box works corectly on my SU650, except for the life thing seems backwards. I have version 3.0.6

 

[agalpha]

It could be doing some background activity, which can cause it to heat up a bit. It will heat when transferring data, when OS writes logs/updates/edits files, or when garbage collection and wear-leveling kick in.

There is no background activity happening. Like I said, it's warm to the touch a little but not hot or not even close to the other drives. The other drives are even warmer to the touch despite showing lower temps.

That may be normal for your SSD but IMO it's still too warm if that is the idle temp.
My standard MX500's (I have 3) all idle at around 30-32C and max out around 53C.
My m.2 SATA III MX500 idles at about 40 and maxes out in the mid-high 50's.
My OS SSD (Samsung 850 500 GB) idles at 29C and I have never seen it above 50C.
But I also have Excellent airflow in my case (2 front intake fans, a side intake fans and a rear and top exhaust fan).
And my MX500 in a USB 3 enclosure idles at 35C and maxes out around 60C under heavy use.

Yeah, I talked to ADATA support again and they say it's completely fine running at these temps.

The A-Data tool box works corectly on my SU650, except for the life thing seems backwards. I have version 3.0.6

Your readings are also wrong. They're showing more then 60000TB of writes and the estimated life isn't backwards. It's supposed to be 100%. Your's is showing the same less life because of the incompatibility with the software.

 

[fzabkar]

Some SSDs have a discrete temperature sensor IC on the PCB whereas others just sense the temperature of the flash controller die. The former will report the air temperature, while the die temperature would be considerably higher.

As for the Total Reads and Writes attributes, I would compare the SMART report before and after writing/reading a 1GB file to/from the SSD. That should help us to make sense of the raw numbers.

Edit: This review attempts to make sense of the SMART data:

https://goughlui.com/2018/12/19/review-adata-ultimate-su650-480gb-3d-nand-ssd-asu650ss-480gt-r/

 

[agalpha]

Some SSDs have a discrete temperature sensor IC on the PCB whereas others just sense the temperature of the flash controller die. The former will report the air temperature, while the die temperature would be considerably higher.

As for the Total Reads and Writes attributes, I would compare the SMART report before and after writing/reading a 1GB file to/from the SSD. That should help us to make sense of the raw numbers.

Edit: This review attempts to make sense of the SMART data:

https://goughlui.com/2018/12/19/review-adata-ultimate-su650-480gb-3d-nand-ssd-asu650ss-480gt-r/

That's the thing bro, The SSD is not that hot to the touch compared to the other drives that are showing a lower temp. I honestly can't figure out this SMART stuff.

 

[fzabkar]

Attribute F2 is the total number of 512-byte sectors read by your OS. When this number reaches 2^32 (32 bits), it rolls over to 0 and starts counting up again. See the SMART calculations by Gough Lui in his review.

Attribute F1 appears to be the total number of 512-byte sectors written to the NAND memory. AFAICT, your SSD compresses the data, so this number will be less than the total number of sectors written by the OS.

Attribute AD could be the min/max/average number of PE cycles. In your case the numbers are 4, 12 (= C in hexadecimal), and 8. This gives you an idea as to the efficiency of the drive's wear levelling.

Your temperature attribute (C2) has 3 components corresponding to the max/min/current temperatures for the current power cycle. These are 64C (= 40 hex), 19C (= 13h), and 56C (= 38h). This temperature is measured on the die of the Maxio MAS09024-B2C flash controller. The other ICs on the PCB are NAND flash memories and buck regulators (for power control). There is no separate temperature sensor IC.

BTW, Maxiotek was the company that JMicron created when it spun off its SSD business.

 

[agalpha]

Attribute F2 is the total number of 512-byte sectors read by your OS. When this number reaches 2^32 (32 bits), it rolls over to 0 and starts counting up again. See the SMART calculations by Gough Lui in his review.

Attribute F1 appears to be the total number of 512-byte sectors written to the NAND memory. AFAICT, your SSD compresses the data, so this number will be less than the total number of sectors written by the OS.

Attribute AD could be the min/max/average number of PE cycles. In your case the numbers are 4, 12 (= C in hexadecimal), and 8. This gives you an idea as to the efficiency of the drive's wear levelling.

Your temperature attribute (C2) has 3 components corresponding to the max/min/current temperatures for the current power cycle. These are 64C (= 40 hex), 19C (= 13h), and 56C (= 38h). This temperature is measured on the die of the Maxio MAS09024-B2C flash controller. The other ICs on the PCB are NAND flash memories and buck regulators (for power control). There is no separate temperature sensor IC.

BTW, Maxiotek was the company that JMicron created when it spun off its SSD business.

So you think thw reason behind the high temps is because the temp is being picked up from the die ?

The smart values are fine from what I could tell with the information given in the above link and provided by you.

 

[fzabkar]

The reviewer at goughlui.com is seeing the same high temperatures as you, so it would appear that this is normal. In any case you should not be comparing your SSD against those that sense the air temperature. For example, Samsung SSDs typically have a separate sensor IC.

There are a few examples of such SSDs in these threads:

https://www.bing.com/search?q=ssd++"temperature+sensor"+site:hddguru.com+fzabkar

 

[agalpha]

I have a WD green SSD running at 38c. I suppose it has a separate sensor ic?

 

[fzabkar]

I don't know, but if you can upload detailed photos of the PCB, with discernible part markings, I could try to identify the components for you.