Dahua NVR5216-4KS2 / NVR5216-16P-4KS2

[SergeyAU]

So now you have the intake at the back and blow the warm air from the psu to your HDD?

its the opposite. the intake is by the hdd. always was, I am just forcing a bit more air through PSU. The exhaust is where the original fan was

 

[Crazykiller]

its the opposite. the intake is by the hdd. always was, I am just forcing a bit more air through PSU. The exhaust is where the original fan was

o i c, you are talking about the psu when you mean inside and outside. Was curious to read. Thought you meant inside the NVR and with outside you meant the fan at the back. Now everything is clear

 

[giomania]

outside is the original fan running at 5v - blowing air IN.
the one on the inside is a replacement 12v 40x40x20mm - Scythe Mini Kaze Ultra. Its blowing air OUT
But as I mentioned previously - you can just run the original fan in its place, just either connect it to 5v connector on the main board or limit the voltage

Thanks.

Since you tested the heat signature of the PSU with no fan, I am also curious if you tested other configurations with your thermal imager, like the original fan setup and your dual-fan setup?

That would be nice to know.


Sent from my iPhone using Tapatalk

 

[SergeyAU]

Thanks.

Since you tested the heat signature of the PSU with no fan, I am also curious if you tested other configurations with your thermal imager, like the original fan setup and your dual-fan setup?

That would be nice to know.


Sent from my iPhone using Tapatalk

As I mentioned before, it will be almost impossible. The part gets pretty hot either way. If I have PSU running without top cover more heat will dissipate than without. And I cannot measure temps with the cover on...
I can only guesstimate that it will be no worse than with just one fan running at full speed, as there are too many ports on the PSU, so the air flow though all components will not be the same.
I did some calculations before, and CFU requirements for that size enclosure was close to the original spec fan - over CFU. All I have done here was to replace 1 fan with 2 fans at half speed on the opposite ends.

 

[giomania]

Took the power supply on the PoE version apart today to see if I can do something about that loud fan. It does not really bother me at the moment, as it sits in the locked metal box in the garage.
You definitely take that loud fan out.
View attachment 19536 View attachment 19537 View attachment 19538
Did not do thickness, but its 20mm. You can check the specs on the fan just by searching by its model number - EFB0412HHD
This is a pic of the actual power supply
View attachment 19539

I removed the fan completely and ran it with 3 cameras connected for 60 min. The hottest part got to around 75C, but I did not put the cover on the power supply or NVR back on, so would probably be hotter.
Don't know what the fan does, but the specs of it say 9 CFM air transfer and about 34dBa (

http://www.delta-fan.com/Download/Spec/EFB0412HHD.pdf

)
View attachment 19540

Thanks SergeyAU - I guess I didn't force it by tugging hard enough and definitely didn't take the PS case off...just removed the unit and couldn't see the fan model number due to the screening. Looking on Amazon, no reviews on that specific model but related model ending in VHD (same size with marginally faster RPM 9000 vs 8200 and airflow 10 vs 9.1) and reviews consistently mentioned noise and a quality control issue - inconsistent terminal wiring (flipping position) on purchase sample of 10. I have to order a WD Purple HD this weekend, so may throw in a fan to experiment with. While I haven't studies in depth, on that caught my eye was the 12v 40x40x20 (careful on replacement not to get the 5v models) Noctua-A4-20 FLX which is Prime and instock. Notably quieter at 14.9dB vs 30.1dB, pushing the same airflow at lower 5000 RPM. It is also a variable for even quieter operation at less airflow and RPM, but don't know if that is user set or automatic demand adjustment.

With an identical size, I cannot imagine the blade differences that would push more airflow at a lower rpm (noise aside). If you do order - please let us know if there is a noticeable difference.
Its hard to guess what is the internal temperature of the psu - dont have any flexible probes to play with it. And what difference that fan makes. Like - can I just take the psu case cover off and run a fan at slow rpm? Or if I run a different fan at lower cfm - how will it imact the temperature? Could not find a datasheet for this particular psu model to see its airflow requirements

Loud fan no more! Got a delivery today - new fan. It only pushes 1/2 the air volume (don't confuse m3/h with CFM - 1 m3/h = 0.6CFM), but so much quieter...Was going to leave it as is, but then I had an old fan in my hand (the loud one). So...
I thought to test it by connecting to the side case fan pins that only supply 5v. And it works! It probably pushes 1/2 the volume, and maybe I could have just left the old one in place and just cut the power in 1/2. But I connected the old fan running on 5v now to the back of the PSU, making sure that it blows IN to the PSU. There is a little screw on the back of the PSU and the long screws that the new fan came with fits the hole and holds the old fan in place.
I had to rotate HDDs a bit, as the thickness of the fan does not allow the drives to stay as they were.
If I was doing the conversion now, I would order 40x40x10mm thick fan (instead of 20mm) either running 5v or 12v and connect it to the back of the PSU.
All in all - great outcome for me. The way it is now, I think it has better airflow than with just 1 tiny fan trying to get all that air through.
View attachment 20152 View attachment 20153 View attachment 20154 View attachment 20155

outside is the original fan running at 5v - blowing air IN.
the one on the inside is a replacement 12v 40x40x20mm - Scythe Mini Kaze Ultra. Its blowing air OUT
But as I mentioned previously - you can just run the original fan in its place, just either connect it to 5v connector on the main board or limit the voltage

So to summarize for future reference:

The original fan inside the PSU is a 12 V, 40 mm x 40 mm x 20 mm Delta EFB0412HHD, with 9 CFM air transfer and about 34 dB(A).

You replaced the original Delta fan inside the PSU with a 12 V, 40 mm x 40 mm x 20 mm Scythe SY124020L Mini Kaze Ultra with 4.86 CFM air transfer and about 19.56 dB(A).

Since this replacement fan theoretically cut the PSU airflow in half, you installed the original fan outside the PSU with 5 V power supply, theoretically cutting the airflow from 9 CFM at 12 V to 3.75 CFM at 5 Volts, with an unknown cut to dB(A).

The theoretical combined airflow of this new configuration would be the Scythe fan's 4.86 CFM + the Delta fans's 3.75 CFM, which is 8.61 CFM total when combined. The result is a much quieter unit, with near-identical airflow in the PSU.

For Noctua fans, the 12 V, 40 mm x 40 mm x 20 mm Noctua NFA4x20 FLX with 5.53 CFM and 14.9 dBa at its fastest speed would not be enough airflow as the original Delta fan inside the PSU.

Alternatively, if you used the 12 V, 40 mm x 40 mm x 10 mm Noctua NFA4x10 FLX with 4.82 CFM and 17.9 dBa on the outside of the PSU as you did above, then the HDD's would fit in their original mounting locations.

The combined airflow with using a dual Noctua fan configuration (NFA4x20 FLX inside the PSU / NFA4x10 FLX outside the PSU) would be 10.35 CFM.

The combined airflow with using the original Delta fan inside the PSU at 5 V to slow it down, and the Noctua NFA4x10 FLX fan outside the PSU would be 8.57 CFM.

I had to convert the Noctua cubic feet per hour to CFM, but am pretty sure the numbers are correct. I guess the only wild card here is the effect of turbulence with two fans basically sandwiched together in a push-pull configuration. I know Noctua uses push-pull configurations in their air coolers, but they have cooling fins in between the fans. This is the reason why I asked if you had performed other thermal tests, and I didn't understand why you could not, but I have never used a thermal imaging device, and just need to educate myself. I will Google that now...

Mark

 

[SergeyAU]

So to summarize for future reference:

The original fan inside the PSU is a 12 V, 40 mm x 40 mm x 20 mm Delta EFB0412HHD, with 9 CFM air transfer and about 34 dB(A).

You replaced the original Delta fan inside the PSU with a 12 V, 40 mm x 40 mm x 20 mm Scythe SY124020L Mini Kaze Ultra with 4.86 CFM air transfer and about 19.56 dB(A).

Since this replacement fan theoretically cut the PSU airflow in half, you installed the original fan outside the PSU with 5 V power supply, theoretically cutting the airflow from 9 CFM at 12 V to 3.75 CFM at 5 Volts, with an unknown cut to dB(A).

The theoretical combined airflow of this new configuration would be the Scythe fan's 4.86 CFM + the Delta fans's 3.75 CFM, which is 8.61 CFM total when combined. The result is a much quieter unit, with near-identical airflow in the PSU.

For Noctua fans, the 12 V, 40 mm x 40 mm x 20 mm Noctua NFA4x20 FLX with 5.53 CFM and 14.9 dBa at its fastest speed would not be enough airflow as the original Delta fan inside the PSU.

Alternatively, if you used the 12 V, 40 mm x 40 mm x 10 mm Noctua NFA4x10 FLX with 4.82 CFM and 17.9 dBa on the outside of the PSU as you did above, then the HDD's would fit in their original mounting locations.

The combined airflow with using a dual Noctua fan configuration (NFA4x20 FLX inside the PSU / NFA4x10 FLX outside the PSU) would be 10.35 CFM.

The combined airflow with using the original Delta fan inside the PSU at 5 V to slow it down, and the Noctua NFA4x10 FLX fan outside the PSU would be 8.57 CFM.

I had to convert the Noctua cubic feet per hour to CFM, but am pretty sure the numbers are correct. I guess the only wild card here is the effect of turbulence with two fans basically sandwiched together in a push-pull configuration. I know Noctua uses push-pull configurations in their air coolers, but they have cooling fins in between the fans. This is the reason why I asked if you had performed other thermal tests, and I didn't understand why you could not, but I have never used a thermal imaging device, and just need to educate myself. I will Google that now...

Mark

Great summary. I could not get Noctua fans cheap enough here, so ordered something similar to test.
I doubt there will be any ill effects in this push-pull config. With so many components inside and so many vent holes around the PSU enclosure extra pressure would just naturally dissipate via the available paths

 

[j4co]

I unpackd my 5416 and powered it on. It sounds like a vacuem cleener..
This one has: psu fan, case out fan and cpu fan on heatsink.

It will go in a technical room in the new house, so i will not try more silent fans yet. Only if it is too much in that room. The case is 1.5 U high so the fans are little larger than the 52xx series.

 

[happf]

I have a non-POE version. The single fan isn't bad. I can't call it quiet, but I won't call it loud either. It is a fast spinning fan. Its going in a hidden area but I could still slightly hear it when I was near it. I want a QUIET nvr.

I removed the top lid, cut out a piece of foam core board to rest ontop, then cut a hole for a 140mm fan over the heat sink. I am running that fan on it's lowest setting. I now have a quiet nvr and using an IR temp gun on the heat sink, my temps have dropped from the low 100s into the high 80s.

 

[giomania]

I have a non-POE version. The single fan isn't bad. I can't call it quiet, but I won't call it loud either. It is a fast spinning fan. Its going in a hidden area but I could still slightly hear it when I was near it. I want a QUIET nvr.

I removed the top lid, cut out a piece of foam core board to rest ontop, then cut a hole for a 140mm fan over the heat sink. I am running that fan on it's lowest setting. I now have a quiet nvr and using an IR temp gun on the heat sink, my temps have dropped from the low 100s into the high 80s.

Pictures?

 

[rado stefano]

Is the IVS skip frame issue fixed???

 

[rufunky]

Is the IVS skip frame issue fixed???

Yes, apears to be fixed with the last firmware update

 

[Sixto]

not fixed with SmartPSS for the Mac.

 

[tigerwillow1]

Interesting how different things work differently. When I upgraded the NVR firmware, the skipped frames problem was mostly fixed on a PC without touching smartPss. The "mostly" part is that single-stepping through an event trigger doesn't work, with smartPss stuck on the last framer preceding the trigger. The video stutters a bit when continuous playing through the trigger but frames are not lost. Doing the same thing via the web interface is now seamless (for me, at least).

 

[c hris527]

not fixed with SmartPSS for the Mac.

Not fixed with smart pss period, view through your NVR directly or the web gui.

 

[3-phase]

I can confirm that my NVR5432 no longer has the missing frame issues. I use hardwired alarms and motion detection. When someone opens my mailbox, a motion event has already been triggered and then a hardwired alarm event happens at the mailbox open. There is a little lag on SmartPSS when it's switching between video files, but no frames are lost. If I export the timeline, it flows seamlessly through multiple events.

 

[Chase]

updates
This is the latest firmware for the NVR5XXX-4K, thanks for sharing.

Index of /DAHUA/OFFICIAL/NVR/NVR5XXX-4K/Latest/

At this time I have the original firmware that was shipped with my NVR. Is this still the link I need to follow for the latest firmware for the 5216-4KS2 NVR?

Is it an easy process to get it updated? Are there instructions anywhere for this?

Is it necessary to upgrade my cameras to the latest firmware as well? I am using the 5231 turrets and the 4231 turret.

 

[tigerwillow1]

At this time I have the original firmware that was shipped with my NVR. Is this still the link I need to follow for the latest firmware for the 5216-4KS2 NVR?

Is it an easy process to get it updated? Are there instructions anywhere for this?

Is it necessary to upgrade my cameras to the latest firmware as well? I am using the 5231 turrets and the 4231 turret.

The link you referenced is what I used. I did the upgrade from the web interface. Instructions? Dream on... I just walked through the setup/system/upgrade menu. I did not do anything with my cameras, and they work as good or better with the new nvr firmware. Easy process? Not totally, but nothing close to some of the typical computer upgrade disasters. The first time I tried the upgrade process rejected the file. Turned out I had a corrupted copy, fixed with a second download. (Hint: The download directory lists the file size. Verify that what you download is reasonably close to this). After the upgrade appeared to be finished and the nvr restarted, I incorrectly thought it was finished and was horrified to see menus in chinese. A couple minutes later it rebooted again and came up working just fine. And of course first go to the info screen to make sure you actually have an older firmware version.

 

[Chase]

I was planning on upgrading from the NVR menu itself and not from the web interface. Is it OK to do so directly from the NVR?

I really only access the NVR from the NVR itself (and not Smart PSS or anything like that)

 

[tigerwillow1]

I vaguely remember hearing that without the web interface you do the upgrade with the .bin file on a usb stick. But if your nvr (like mine) has only one usb port, it looks like a dilemma because you need a mouse plugged into the usb port. Maybe a usb hub works? I don't know and would like to learn from somebody who does know.

 

[c hris527]

I vaguely remember hearing that without the web interface you do the upgrade with the .bin file on a usb stick. But if your nvr (like mine) has only one usb port, it looks like a dilemma because you need a mouse plugged into the usb port. Maybe a usb hub works? I don't know and would like to learn from somebody who does know.

I just upgraded 2 of my Customers NVR's and both NVR's had two usb ports, doing it directly through the nvr would NOT work. I called my usa reseller and he said to try it through the web gui and it worked...go figure. The issue I was having was if I had a camera set on motion, it would stop recording all together if the nvr rebooted.