Norco Rackmounts: Issues & Improvements


PhatalOne

Recommended Posts

Norco 4020 and 4220 cases are quite popular picks for setting up a high density unRAID server since they are comparable to buying several high end 5in3 devices but give you an entire rackmount case to work with.  However, both of these cases have had their own unique issues when it comes to effectively meeting the needs of the discriminating unRAID user.  As such, I'm setting this thread up to discuss issues we have with these cases and for people to also post their solutions.  This thread is not just for 4020 and 4220 owners, if you have any Norco rackmount case, have issues with it, or have made improvements to it, please feel free to post. Together, we can identify and resolve many of these issues and make these case better for our uses. 

Link to comment

Suggested Future Improvements

 

Cooling

1.) 120mm fans or at least a choice of fans, and a temperature-based fan speed controller.

2.) A full sized 3.5" bay for a fan controller/multifunction LCD panel OR some fan controls integrated into the front panel.

3.) Side fan mounts (1 each side for 120mm) for extra exhaust.

4.) Thermal diode on each drive bay backplane with LED/alarm and selectable temp indication (i.e. yellow at 45C and red at 55C, using jumpers on the backplane).

5.) Remove the random vent holes.

6.) Blanking plates or foam inserts for empty slots to stop air being drawn through them as path of least resistance

7.) Removable/washable dust filters

6.) With a trayless system,  make the front panel a screen filter door for better airflow.

 

Drive Bay

1.) Trayless drive bays. (Or at least trays that don't need screws... such as having pins sticking up from the bottom of the tray that fint into the screw holes in the bottom of the drive.)

 

General Construction

1.) Multiple front panel USB ports.

2.) Removable motherboard tray.

3.) Better quality front and rear mounting rails.

3.) An improved lid lock that is more than just a springy metal clip.

4.) Replace the membrane switches with proper switches.

 

 

 

Norco RPC-4220 Possible Issues

1.) Some SAS backplane connectors are located to close to the metal support beams and locking connectors with metal clips cannot lock into place.

2.) The SAS/SATA backplanes have some degree of flex to them and in their original state the SATA connectors may not align with the drive connectors.

 

 

If you have any ideas for improvements not on this list, throw them out there.  The goal is to have to good ones stick and the bad ones drop off.

Link to comment

Common Issues - RPC-4020 & RPC-4220

 

1.) Stock fan plate is loud and has design issues that prevent the fans from operating at peak efficiency.

 

Solution: Several users have created their own fan plate modifications to replace the stock fans with larger and quieter fans.

 

RPC-4020 Custom Fan Bracket

http://lime-technology.com/forum/index.php?topic=2935.msg26161#msg26161

 

RPC-4220 Custom Fan Bracket

http://lime-technology.com/forum/index.php?topic=2935.msg34584#msg34584

 

2.) The performance of the 80mm fans in the motherboard area is important factor in the effectiveness of the fan plate.

Link to comment

The Norco is an excellent case with high build quality for the price. The concept is very good and it is unique in the market (again for the price). However, and this is important, if you are running 20 disks it is not fit for purpose other than for a high end home user and without careful management you can easily cook your drives out of warranty.

 

This is a real shame as I am certain that with some thought and experimentation their design could solve this issue.

 

Bullet point list of potential improvements:

 

Trayless disk mounts allowing for better interdisk ventilation

Replace the cheap rubbish nipple switches with proper switches

Remove the random vent holes

Present the power for the backplane beside the PSU neatly

Present the SATA port on the MB side of the fan plate

BETTER FRONT AND BACK MOUNTING RAILS - the rails supplied simply do not work with a loaded server and will fail catastrophically i.e. your server would fall out of the rack

Blanking plates/foam inserts for empty slots to stop air being drawn through them as path of least resistance

removable/washable dust filter in the caddies and any other intake vents

Lid lock that isnt just springy metal clip (try and get this thing open when one of them gets dented its a 2 man job with hammers)

Option for simple LCD display

removal MB tray

multiple front mounted USB ports

 

I would like to see caddies abandoned completely in favour or 20 in situ frames that you could slide the drive into and clip shut with some sort of spring clip. Then the whole front of the server could be a screen filter door. this would allow for maximum airflow fasy each disk.

 

Link to comment

Bottom line, the air pressure in the motherboard area is very important to the performance of the fan plate.

 

This is why I want an exhaust fan mount on each SIDE of the case, to increase exhaust.  It may not help much in a rack room with multiple racks packed in tight side-to-side, but with a single rack, it will make a world of difference.

 

bubbaQ & NAS,  I've been looking at possible solutions to relieving the positive pressure in the motherboard chamber. 

 

Here are some permanent solutions that would be quiet and would not require any software involvement in managing the fans:

 

1.) Create some ducting with larger fans that would exhaust out the 2 80mm fan brackets.

I envision this going one of two ways:

a.)The duct work is attached to the fan plate and at the 80mm exhaust outlets and helps shape/guide the airflow along with some 80mm fans to lower the static pressure of the duct.

b.)Create some duct work that moves the exhaust fan mounts to a different location so that two mid-to-high flow 120mm fans could be used. 

 

The first potential downside I see with either design would be the design of the duct work is crucial and if done improperly could interfere with the cooling of system components.

 

2.) Dremel a cutout in the top of the case for a 200-250mm exhaust fan.

One of these fans would exhaust 100+ CFM all while not even breaking 22dB.  Combined with some quiet 80mm fans and the motherboard area should be pressure neutral (if not negative).  Obviously, if the server hangs out anywhere but the top of the rack, this is not going to work.  Additionally, it is not for the faint of heart since it requires attacking your case lid with power tools. Lastly, I've yet to even find extra aftermarket grilles/guards to cover any exposed fan blades.

 

3.) Some combination of 1 and 2 where (an) exhaust port(s) is/are created in the case with duct work shaping air from the fan plate to the exhaust port(s)

 

I'll update this post with some possible software solutions I can entertain later.

Link to comment

I've had some more time to think and came up with this option:

 

Get a 25-50w rheostat such as this one:

http://www.sidewindercomputers.com/si50warh.html

 

 

It would need a knob or something (which they just happen to conveniently sell):

 

http://www.sidewindercomputers.com/ststknforsi2.html

http://www.sidewindercomputers.com/poknforsi25w.html

 

Mount the rheostat with the knob piece sticking through the rear grating, attach the knob to the back side of the case, plug in your ultra high powered crazy fans to it, and then control the power voltage to the fans manually as needed.  If you use fans like the Delta EHE's or the tornado's I would suggest sealing the back of the case and creating some sort of shield or simple duct for the 80mm fans to keep the PSU from becoming an intake vent since, at full power, they will suck air in from the outside instead of pulling it through the case.

Link to comment

I like the plan but can i suggest that since it relies on a human being around to deal with the fan speed in the event of an overheat it is far from ideal.

 

Throwing some ideas in:

 

Adding fans to the case lid would be easy. Obviously it would depend on cabinet placement but there is room for many fans. Rather than controlling fan speed can we turn on an ever increasing amount of fans as required (or is that essentially the same problem)?

 

Are there such thing as case fans with tunable thermistors?

Link to comment

NAS,

 

The other side that I hven't looked at fully is to create some combination of software that does the following:

 

1.) Monitors SMART information from the disk drives

2.) User configurable temperature limits & fan activity percentages

3.) When temperature reach thresholds on any SMART probe, increase the fan speed on specified fan headers to increase exhaust flow.

4.) When temperature stays below threshold for user defined amount of time, decrease fan speed to previous level.

 

The tricky part about this is that your motherboard must to be able to be control fan speeds using the fan headers. (not all do)

 

I know some fan controllers have profiles like this setup based on an individual temperature probe, but obviously that is not going to help if the probe is on an inactive drive or a cool spot in the case.

 

I'll be looking into this more this evening, but the increased number of variables means an increased number of things to break in the software as well.

 

 

Link to comment

You might just try one of these fans:

http://www.frozencpu.com/products/7545/fan-444/SilenX_120mm_x_25mm_14dBA_72CFM_Ixtrema_Pro_Thermistor_Fan_IXP-74-14T.html

http://www.frozencpu.com/products/7543/fan-442/SilenX_80mm_x_25mm_14dBA_32CFM_Ixtrema_Pro_Thermistor_Fan_IXP-54-14T.html

 

They have a thermistor you simply place near the heat-producing disks and when it gets hotter, the fans speeds up.  When cooler, it slows down.

 

The SilenX fans come in various sizes too.  I use one in my media rack, powered by a wall-wart, blowing air under the Cable-TV SA8300HD box since it runs so warm.  I just tucked the thermistor between the units... works just fine.

 

(I have nothing to do with the link posted in this message, it was just the first that came up in a Google search)

 

Joe L.

Link to comment

Joe, that is a pretty good find. I had looked at Silenx in the past but never paid attention to that functionality... it would be nice if we could find fans similar to the powerful Delta fans with that feature.

 

The only other answer I can come up with is to buy a fan controller that allows for user adjustable temperature controls such as:

 

http://www.performance-pcs.com/catalog/index.php?main_page=product_info&cPath=87_137&products_id=23822

 

That particular unit takes 2 5.25" bays so it isn't an option for the Norco unless you do some heavy modding.  I'm sure they make some smaller units with similar functionality, but I don't know of any for the low profile bays we have.

Link to comment
Intersting i never considered that. I assumed that made no difference since many cases come with the PSU in a compartment of its own with no access to normal case ventilation

 

If the PSU compartment is self contained, with it's own private intake vented to the outside of the case and own private exhaust to the outside of the case, then it is OK.  Otherwise creating a negative pressure in the inside of the case where the intake fan for the PSU is, will quickly overheat the PSU.

Link to comment

NAS,

 

To go after the lowest hanging fruit (seriously beefing up the rear fans), here is what I think would need to be done to ensure airflow is directed through the case:

 

1.) Seal every opening on the back and side of the case.

2.) Seal the edge between the case and the top lid.

3.) Create a partition to make it near impossible for the case fans to suck air through the PSU fan and vents.

 

The downside of this that I see is that you are still limited to two 80mm fans unless you do some serious modding of the back panel.  To get sufficient airflow (~160 cfm) you'd need some very powerful and loud fans (at least 50dB's each).

 

In this instance, after you've put in all the time and effort of properly sealing the case, I think it would be better to get some high flow 120's and create a new mid mount fan plate. By getting some high flow 120's you'd get decent static pressure which would help exhaust air from the positive pressure area around the motherboard and PSU and the noise would be a bit lower.

 

In the 120mm plate scenario, you may not need any rear 80mm fans depending on the static pressure of the 120mm fans in the mid mount plate.  The reason I say this is that for the 80mm fans to have any impact whatsoever, their static pressure generated from the 80mm fans must be greater than the static pressure generated by the 120mm fans at 1ft or 30.5cm. This part is very hard to measure for the average user. For example, I know the 80mm Deltas in the 4220 plate created a very noticeable plume of air pressure behind them that extended for at least a foot.  What the static pressure was along the various points of that plume, I have no clue. (I'm pretty sure I could go home and calculate it based on the laws of physics, but I haven't been that interested in doing it yet.) I do know that, if 120mm are moving more air than at that distance from the fan plate than the 80mm fans would are rated for, the 80mm fans are basically doing nothing but impeding air flow through the case.

 

I also want to be clear that I'm basing this part off of my very limited understanding of fluid dynamics/physics.  If anyone notices somewhere where I am wrong, please correct me.  My dive into creating the new fan plate for the 4220 really connected the concept of static pressure vs. CFM that BubbaQ stressed in the other thread. 

 

BubbaQ Do I have this concept right: Basically, it really doesn't matter how much CFM the fan is capable of generating. If the static pressure of the void you are trying to exhaust air into equals the static pressure that can be generated by the fan, then the fan effectively moves 0 CFM.

 

 

Link to comment

No... .your complete description of static pressure (head) is wrong.  Air is a FLUID and all the normal rules of fluid dynamics apply.  Air is FLOWING in a case so static head is irrelevant... the environment is dynamic, even at steady state conditions.  Fan performance can be greatly affected by NPSH (net positive suction head).  If you have a positive pressure in the second chamber, those to 80mm fans will significantly exceed their rated flowrates.

Link to comment

Interesting, now you've made me interested enough to go brush up on my physics... for some reason that really don't care to teach that stuff for Accounting / Finance. It's quite funny how you end up having to relearn things that, as a kid, we never thought we would use in life.

 

Bubba, in my previous example I asked you about to confirm my understanding or lack thereof, I had pictured a sealed box with a two equal fans on either side blowing air into it.  In this example, since the static pressure exerted by each fan into the void is equal, my elementary understanding was that they would effectively do no work.  With that clarification is my understanding still wrong?

 

Now, in the instance of air moving from a high pressure zone to a low pressure zone, I see, and agree with your point.  However, could you please clarify for my understanding on the multi-fan example based on some real world numbers.  I will attempt to illustrate with bad ASCII art.

 

 

   <<||         

   <<|| Fan1 (20CFM, 1.21 mm/H2O)

   <<|| (rear case mount)

 

  <<<<<<<<<|||

  <<<<<<<<<||| Fan2 (105CFM, 34.5 mm/H2O)

  <<<<<<<<<||| (fan plate)

 

In this instance shown above, Fan2 sits some distance behind Fan1.  Due to the drastically different capabilities of the fans, I believe there would be a dividing line somewhere in the exhaust plume of Fan2 where placing Fan1 would not provide any additional benefit because the speed of the fluid is still dictated by Fan2. 

 

Basically, If Fan2 is doing the same or greater amount of work at any point along it's exhaust plume where Fan1 is mounted, then Fan1 cannot provide any additional benefit since the fluid is already moving as fast or faster than Fan1 can move it.  If I'm still off base could you provide me a few topics to brush up on tonight? 

Link to comment
Due to the drastically different capabilities of the fans, I believe there would be a dividing line somewhere in the exhaust plume of Fan2 where placing Fan1 would not provide any additional benefit because the speed of the fluid is still dictated by Fan2.

 

Nope.  Friction losses is what you have to consider, not relative speed.

Link to comment

Due to the drastically different capabilities of the fans, I believe there would be a dividing line somewhere in the exhaust plume of Fan2 where placing Fan1 would not provide any additional benefit because the speed of the fluid is still dictated by Fan2.

 

Nope.  Friction losses is what you have to consider, not relative speed.

 

Accounting for friction losses and other variables, is there not some range along the exhaust plume of Fan2 where it no longer not makes sense to add Fan1?

Link to comment

I'll do some more research tonight to refine my terminology. The reason I used the phrase "exhaust plume" was more of an illustration of the noticeable difference in air pressure at various points behind two different types of fans when they are tested in a free air environment. 

Link to comment

BubbaQ,

 

Here is what I have gathered so far, please correct me if I am wrong.

 

1.) We are now working with a system that has both static and dynamic pressure. 

2.) From the research that I've done, I believe I have a good grasp on static pressure.  I believe that one of my original assumptions (below), though worded poorly was correct. I've included an example to make sure I understand static pressure.

 

BubbaQ Do I have this concept right: Basically, it really doesn't matter how much CFM the fan is capable of generating. If the static pressure of the void you are trying to exhaust air into equals the static pressure that can be generated by the fan, then the fan effectively moves 0 CFM.

 

Example 1: Calculating Static Pressure of a Closed Space

Assume the motherboard area is 1 ft3, there is no leakage, standard air pressure is 10,000 mm H2O, and the static pressure of the fan is 34.5mm (Delta 80mmx38mm fan).  Therefore:

 

10,034.5 mm H2O * 1 ft3 / 10,000 mm H2O = 1.00345 ft3 pressurized volume

 

Once, the space reaches its pressurized volume, there is no longer any airflow into the space because the fan static equals the pressure of the space.

 

 

Example 2: Calculating Leakage Rate

Assuming the fans from above create a pressure increase in zone 1 of 34.5mm H2 and have a net flow rate of 320CFM here is my leakage calculation.  (This part I'm sketchy on and could be very wrong.)

 

Leakage (CFM free air) = (V * (P1-P2)/T * 14.7 psi) * 1.25

 

V = 1 ft3

P1 = 1.358268 psig (34.5mm H2O * .03937 in./mm. = 1.358268 psig)

P2 = 0 psig

T = 1 minute

14.7psi = 1 atmosphere (free air)

1.25 = Constant to correct to normal system pressure

 

1 ft3 * (1.358268 psig - 0 psig) / 1 minute * 14.7psi) * 1.25 coefficient = ~25CFM Leakage (free air)

 

I have a hard time believing this calculation is entirely correct since the formula came from a guide on compressed air systems. (which our is most decidedly not)  In my estimation, the constant should be higher because the pressure in the system never decreases like it would in a compressed system with a leak. I will look for a better equation tomorrow and dig a bit deeper into the other areas I haven't had a chance to review. Could you give me some feedback based on the work I've shown above.  Personally, I believe the gradient from the pressure difference inside of the case to the free air environment outside the case should promote a higher transfer rate than 25CFM when the fans can deliver 320CFM.

 

Is this example correct?

 

That's it so far. I'm still working on things and haven't even touched dynamic pressure and a host of other topics.... you'd almost think one could make a career out of it. ;)

Link to comment

Why are you dealing with a closed space?  This is about cooling a computer case, with dynamic, but steady state, fluid flow.  NOTHING is static. 

 

You have 2 chambers.  The front (with negative pressure) has restricted intake and the fan plate for exhaust into the second chamber, and then the PSU and other exhaust fan evacuating the second chamber (with positive pressure) to the outside.

 

Now unless the front intake is GREATLY restricted, or the exhaust fans in chamber 2 are extremely powerful, the second chamber will have positive pressure.  That positive pressure will substantially increase the performance of the exhaust fans.  But the fact there is any positive pressure in the second chamber, hurts performance of the fans moving air from chamber 1 to 2.

 

This is why I posted many moons ago that the fans on the fan plate need to be selected for both airflow AND high deltaP, and not just on airflow, and the exhaust fans need to be sized for only for high airflow.

 

If you beef up exhaust on chamber 2 to create negative pressure in chamber 2, you will have kickass cooling, but you will fry the PSU since the measly fan in the PSU will be starved.  In that case, you have 2 choices.  1) you can move the PSU further into the case, so the PSU intake and exhaust are BOTH in chamber 2.... PSU fan will perform to spec and the air from the PSU exhaust will then exit the case through the exhaust fans, or 2) put the PSU in its own chamber with a separate intake.

 

Static head measurements of a fan are worthless pieces of info for this task.  Pressure vs airflow is 2-dimensional a curve.  2 fans can both deliver 30 Pa of pressure at zero flow (static), but at 50 cfm, one produces 20 Pa and the other only 5 Pa.

 

Plus if there is a positive pressure (on the fan suction side), that is a FLOW, not static, ... so if you have 10 Pa of positive head, and the fan delivers 20 Pa of pressure at 50 CFM, you can NOT add the 2 and predict the fan will produce 30 Pa of output pressure at 50 CFM.  Fan performance curves change, sometimes dramatically, with different values of head..

Link to comment

You might just try one of these fans:

http://www.frozencpu.com/products/7545/fan-444/SilenX_120mm_x_25mm_14dBA_72CFM_Ixtrema_Pro_Thermistor_Fan_IXP-74-14T.html

http://www.frozencpu.com/products/7543/fan-442/SilenX_80mm_x_25mm_14dBA_32CFM_Ixtrema_Pro_Thermistor_Fan_IXP-54-14T.html

 

They have a thermistor you simply place near the heat-producing disks and when it gets hotter, the fans speeds up.  When cooler, it slows down.

 

The SilenX fans come in various sizes too.  I use one in my media rack, powered by a wall-wart, blowing air under the Cable-TV SA8300HD box since it runs so warm.  I just tucked the thermistor between the units... works just fine.

 

(I have nothing to do with the link posted in this message, it was just the first that came up in a Google search)

 

Joe L.

 

Just as an FYI alot of the specs of this brand of fans have been debunked. Whilst I suspect the backlash is OTT and the fans are quiet and strong they almost certainly dont match the specs. Also there are countless posts about them failing.

 

If you are bored you can start reading from here:

 

http://www.ocforums.com/showthread.php?t=431776&highlight=SilenX

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.