-
Important Announcement for the TrueNAS Community.
The TrueNAS Community has now been moved. This forum will now become READ-ONLY for historical purposes. Please feel free to join us on the new TrueNAS Community Forums
Norco RPC-4224 and hard drives that are too hot
- Thread starter cyberjock
- Start date
- Status
Ri-Vier / Richard
Cadet
- Joined
- Oct 22, 2013
- Messages
- 6
To get the temperature down on Norco cases you can try to open up the holes in the disk tray with a small screw driver. If open up, the air can flow better at the harddisk.
I have no idea what you are doing with that screwdriver, but breaking your caddies won't work out for you. If you read online somewhere to open the holes, some of the older models that are 2+ years old have metal slats that you can slide open or closed. They are "supposed to be closed if the caddy is empty and open if there is a drive installed. I do have that configuration, and all of mine are open.
But the problem with the Norco cases not having better cooling has to do with the limited airflow curve versus the differential pressure across the fan and the very high density of the disks.
So I fail to see the solution you are offering. Can you elaborate?
But the problem with the Norco cases not having better cooling has to do with the limited airflow curve versus the differential pressure across the fan and the very high density of the disks.
So I fail to see the solution you are offering. Can you elaborate?
Ri-Vier / Richard
Cadet
- Joined
- Oct 22, 2013
- Messages
- 6
Hi Cyberjock,
I know you can open en close the metal slide, but if you have trays as the picture you can try to open up the front of the tray to remove some of the plastic ribbons. That way you open up more space to let more air flow pass the disks. This is a tested method by other customers and this bring down the heat.
Of course you can use the Ri-vier own brand chassis where the trays are more open up like the RV4324-03A and has proven better cooling than the Norco's.
Those chassis has also better and stable backplanes than the Norco chassis and the backplanes are made and tested by Intel.
I know you can open en close the metal slide, but if you have trays as the picture you can try to open up the front of the tray to remove some of the plastic ribbons. That way you open up more space to let more air flow pass the disks. This is a tested method by other customers and this bring down the heat.
Of course you can use the Ri-vier own brand chassis where the trays are more open up like the RV4324-03A and has proven better cooling than the Norco's.
Those chassis has also better and stable backplanes than the Norco chassis and the backplanes are made and tested by Intel.
Ri-Vier / Richard
Cadet
- Joined
- Oct 22, 2013
- Messages
- 6
Hi Cyberjock,
Ri-vier sell Norco's only in the EU.
Other chassis can be sold outside the EU (contact Ri-vier), but the shipment cost are very high, and therefor not interesting, sorry.
Ri-vier sell Norco's only in the EU.
Other chassis can be sold outside the EU (contact Ri-vier), but the shipment cost are very high, and therefor not interesting, sorry.
Boo, that's disappointing. Unfortunately, most EU people try to stick to the tried and true cases because they can and do pay VAT taxes. Here in the USA someone would normally buy one just to see how well it works as many of us aren't going to spend money on a case to "find out" how well(or how poorly) it works. Guess we'll have to wait and see who is willing to buy the first one and find out how good it is. :(
I just sent you a PM.
I just sent you a PM.
jgreco
Resident Grinch
- Joined
- May 29, 2011
- Messages
- 18,680
OMG THERE'S A BOTTOM TO THE TRAY?!?!
No wonder these stupid things toasty their drives. Did you manufacturers learn NOTHING from AIC's mistakes?
In a 4U enclosure you have 7 vertical inches and 6 rows of drives. A typical 3.5" drive is just over 1" high (ST4000DM000 is 1.03" for ex). So you have 6.18" used. The chassis itself is supposed to be just shy of 7" (usually about 6.95"). So you have just 3/4ths of an inch to hold the top plate, bottom plate, and any interior structure.
In an imaginary world where the chassis was made of zero thickness Unobtanium (2d plane!), you are still left with a cross-sectional area of only .77" x 17" through which the air can flow, or about 13 square inches. Which'd be great if it was a free and clear rectangular duct, but it is actually a mostly-obstructed duct (since 7" x 17" = 119 sq - 91% blockage). So even in our imaginary world of optimal chassis, ... bad.
But in the real world it is worse. Even in the current Supermicros, I'm skeptical if anyone claims more than 1mm clearance above and below a drive (individually enclosed in a grid). But visual estimation does suggest something close to that, so conservatively .75mm times 6 rows times two (above and below) is 9mm (guesstimate). Sticking in a bottom to the trays would effectively eliminate about half of that, I'd guess?
So I'm kind of horrified. I hate the bottomless trays but they're sensible from a strictly engineering PoV. I keep reading about the difficulties the Norco guys have. I continue to come to a better understanding of why that is.
And metal slats to block airflow in blank slots? Use a plastic filler tray. Sufficiently solves the problem a tray bottom does, AND blocks the air without chance of some ops person who is unfamiliar with the chassis not realizing there's a magic slat to open on the tray to enable airflow.
I'm sorry cj 'cuz I know you like your Norco but boy ... sure seems like it wasn't designed by people who had looked at the competition to figure out what worked and failed. And I say that comfortably because AIC made a lot of stupid choices which should have been easily-learned-from. And we sold AIC back in the day. heh
No wonder these stupid things toasty their drives. Did you manufacturers learn NOTHING from AIC's mistakes?
In a 4U enclosure you have 7 vertical inches and 6 rows of drives. A typical 3.5" drive is just over 1" high (ST4000DM000 is 1.03" for ex). So you have 6.18" used. The chassis itself is supposed to be just shy of 7" (usually about 6.95"). So you have just 3/4ths of an inch to hold the top plate, bottom plate, and any interior structure.
In an imaginary world where the chassis was made of zero thickness Unobtanium (2d plane!), you are still left with a cross-sectional area of only .77" x 17" through which the air can flow, or about 13 square inches. Which'd be great if it was a free and clear rectangular duct, but it is actually a mostly-obstructed duct (since 7" x 17" = 119 sq - 91% blockage). So even in our imaginary world of optimal chassis, ... bad.
But in the real world it is worse. Even in the current Supermicros, I'm skeptical if anyone claims more than 1mm clearance above and below a drive (individually enclosed in a grid). But visual estimation does suggest something close to that, so conservatively .75mm times 6 rows times two (above and below) is 9mm (guesstimate). Sticking in a bottom to the trays would effectively eliminate about half of that, I'd guess?
So I'm kind of horrified. I hate the bottomless trays but they're sensible from a strictly engineering PoV. I keep reading about the difficulties the Norco guys have. I continue to come to a better understanding of why that is.
And metal slats to block airflow in blank slots? Use a plastic filler tray. Sufficiently solves the problem a tray bottom does, AND blocks the air without chance of some ops person who is unfamiliar with the chassis not realizing there's a magic slat to open on the tray to enable airflow.
I'm sorry cj 'cuz I know you like your Norco but boy ... sure seems like it wasn't designed by people who had looked at the competition to figure out what worked and failed. And I say that comfortably because AIC made a lot of stupid choices which should have been easily-learned-from. And we sold AIC back in the day. heh
Has anyone tried the RV4324-03A? I keep getting PMs in IRC and whatnot asking if I know how good that case is and nobody seems to have actually used one. Anyone broken the ice on this case and can vouch for how well it works?
I looked at buying one(I'm in the market for a new 4U rack case) but the cost with shipping to the USA and a discount from @ri-vier-richard makes it something that's not viable due to cost. Its almost as expensive as a Supermicro case.
It seems there is a potential customer base, but nobody wants to be the guinea pig. I'm really wishing I could get one, but its so darn expensive being across the pond. It pretty much would have to be donated to make it cost effective to try it. Of course, if it works well they could probably make their money back quickly as there's a definite demand.
Anyway, anyone try one or knows someone that did?
I looked at buying one(I'm in the market for a new 4U rack case) but the cost with shipping to the USA and a discount from @ri-vier-richard makes it something that's not viable due to cost. Its almost as expensive as a Supermicro case.
It seems there is a potential customer base, but nobody wants to be the guinea pig. I'm really wishing I could get one, but its so darn expensive being across the pond. It pretty much would have to be donated to make it cost effective to try it. Of course, if it works well they could probably make their money back quickly as there's a definite demand.
Anyway, anyone try one or knows someone that did?
Ri-Vier / Richard
Cadet
- Joined
- Oct 22, 2013
- Messages
- 6
Hi Cyberjock and others,
I get a new line of cases (expecting in February) and the 24 bay is this:
https://ri-vier.eu/rivier-4u-24bay-sassata-storage-chassis-rvs406a-p-352.html?cPath=1_3_7
other RVS series are these:
https://ri-vier.eu/rivier-3u-16bay-sassata-storage-chassis-rvs306a-p-351.html?cPath=1_3_6
https://ri-vier.eu/rivier-2u-12bay-chassis-with-sas-backplane-rvs206a-p-323.html?cPath=1_3_5
https://ri-vier.eu/rivier-2u-8bay-chassis-with-sas-backplane-rvs106a-p-324.html?cPath=1_3_5
I think I have find a shipper who can ship one carton for 110 EUR EX VAT to the USA.
Is there any interest in this case or another of the RVS series?
Also we are seaching for one or more distributors for our cases in de USA so the shipping prices can be better. We have good reseller prices. So if you are one, please contact me.
Richard.
I get a new line of cases (expecting in February) and the 24 bay is this:
https://ri-vier.eu/rivier-4u-24bay-sassata-storage-chassis-rvs406a-p-352.html?cPath=1_3_7
other RVS series are these:
https://ri-vier.eu/rivier-3u-16bay-sassata-storage-chassis-rvs306a-p-351.html?cPath=1_3_6
https://ri-vier.eu/rivier-2u-12bay-chassis-with-sas-backplane-rvs206a-p-323.html?cPath=1_3_5
https://ri-vier.eu/rivier-2u-8bay-chassis-with-sas-backplane-rvs106a-p-324.html?cPath=1_3_5
I think I have find a shipper who can ship one carton for 110 EUR EX VAT to the USA.
Is there any interest in this case or another of the RVS series?
Also we are seaching for one or more distributors for our cases in de USA so the shipping prices can be better. We have good reseller prices. So if you are one, please contact me.
Richard.
I'm interested, but the problem is after shipping and everything its so expensive that I could easily just get a Supermicro chassis for a little more. It turns into a problem because when the cost is the same I can go with a tried-and-true Supermicro or your design. It's just a simple question of which is less risky when the price is the same. Obviously nobody wants to buy a case that won't work for them, so why go with an untested case when you can get a Supermicro. :(
jgreco
Resident Grinch
- Joined
- May 29, 2011
- Messages
- 18,680
That looks disturbingly like Chenbro's product, though it is hard to tell on a cellphone. On a scale, the Norco stuff is near the bottom (2 out of 10 maybe), AIC and Chenbro are rated a bit better (4/10), Supermicro offers best general options (7/10) and from there you move on to stuff by the big storage mfrs which isn't readily available for FreeNAS type systems.
Scharbag
Guru
- Joined
- Feb 1, 2012
- Messages
- 620
Max temp of my 13 3TB drives in a Norco 4220 case has been 37C. That is even when all of them are scrubbing and the server is in a warm utility room. I have the stock 4@80mm fans in there. So far it has been a pretty good case. Perfect world, I would get a SuperMicro.
That is pretty close to what I am experiencing. I have seen about 34C as the max during a scrub, but my server is in a basement which is much cooler. All the drives seem to be right around the same average temperature also.
sorry to dig old thread. Is this fan good for this particular case? http://www.newegg.com/Product/Product.aspx?Item=N82E16835129055
I use these fans currently and they work great, I would highly recommend them over any 120mm fan. Just make sure you have a fan controller because they are VERY loud at full speed.
View: http://www.amazon.com/Supermicro-FAN-0126L4-Assembly-NIDEC-V80E12BHA5-57/dp/B00OJ643L8/ref=sr_1_3?ie=UTF8&qid=1413822324&sr=8-3&keywords=FAN-0126L4
- Joined
- Feb 15, 2014
- Messages
- 20,194
Those are as good as useless in a 4U chassis. 120mm fans aren't the best choice for a 24-bay 4U chassis, much less those things. You need some proper thick 80mm fans.
- Joined
- Feb 15, 2014
- Messages
- 20,194
You need industrial fans, basically. Like the Supermicro-branded fans above. You can also shop around for Sanyo-Denki or Delta fans and assemble something yourself.
Nope. If you read all this data you'll find that Ericloewe's response is basically spot on. You'll also find that 80mm fans perform significantly better than 120mm fans.
- Status
