Overclock.net banner

1 - 20 of 38 Posts

·
Registered
Joined
·
243 Posts
Discussion Starter #1
Hey guys,

I am curious on how the EK Vardar 120ER EVO operate. Especially how they ramp up in speed once powered up.
I have 12 EK Vardar ER (non EVO from 2017) that start at full speed even if the PWM cycle is at 50% and then slow down to the PWM setting. It is pretty annoying and usually trip the overcurrent protection on the fan controller (when tying up 6 on an AQ6 channel).

Do the EVOs do the same?

Thanks
 

·
Premium Member
Joined
·
2,390 Posts
Yep mine do. Only goes for a few seconds before the AQ kicks in

Never had them trip the over voltage and I have 6 per channel as well.
Even have a pair of 6 on a Quadro
Using splitty9’s?
 

·
Registered
Joined
·
243 Posts
Discussion Starter #3
Yep Splitty's. Support told me that the controlling chip may be damaged in the AQ and this is usually not covered by warranty (because of bad behaving fans - go figure). So I haven't RMAd it.

So, you say that your EVOs ramp up at full speed when powering on even though the PWM duty cycle is not at 100%?

Another thing: AQ kicks in well before sending voltage to the fans. You can confirm the behavior by unplugging and replugging the fan while it's running (it will start at 100% and drop down to the PWM setting).

(note: it's overcurrent not overvoltage)
 

·
Premium Member
Joined
·
2,390 Posts
Yep all of mine do that. Least I think they do. Don’t have my rig up and running atm so sorry for the late reply.
I’ll test it out next week when I get it running again but as far as I can remember they start up at 100% then drop down to the AQ settings after a few seconds.
 

·
Registered
Joined
·
243 Posts
Discussion Starter #5
Thanks for the info. Appreciate it.

If that's the case, they're not for me as the ones I already have do the same. I measured over 3 Amps for 6 fans connected so that is bound to create issues sooner or later (as it did with my AQ6). I moved them on the Quadro, but I expect it to fail the same way down the lane (AQ6 failed 2 days after 2 years).
 

·
Registered
Joined
·
323 Posts
Hey guys,

I am curious on how the EK Vardar 120ER EVO operate. Especially how they ramp up in speed once powered up. I have 12 EK Vardar ER (non EVO from 2017) that start at full speed even if the PWM cycle is at 50% and then slow down to the PWM setting. It is pretty annoying and usually trip the overcurrent protection on the fan controller (when tying up 6 on an AQ6 channel).

Do the EVOs do the same?

Thanks
Just wanted to say using an aquaero 6 my evo fans do not ramp up when they start spinning which they shouldn’t at least. Now my ER fans I have 3 on one radiator that don’t ramp up. They can stop spinning and then start back up spinning very slowly. Another 2 ER fans ramp up and the only way around it that I found was to set a low minimum pwm % like 10 percent for those fans, or set them as speed or power (voltage control) and they didn’t ramp up. Personally I stuck with the minimum pwm % since they spin at about 300rpm.
 

·
Registered
Joined
·
243 Posts
Discussion Starter #7
Another 2 ER fans ramp up and the only way around it that I found was to set a low minimum pwm % like 10 percent for those fans, or set them as speed or power (voltage control) and they didn’t ramp up. Personally I stuck with the minimum pwm % since they spin at about 300rpm.
That's exactly how my fans operate and how I managed to work around this issue. But, the problems I am experiencing with AQ6 are at cold boot unfortunately.

It seems that they have multiple iterations for these fans and this behavior is not mentioned anywhere. Not even the EK support people know about this. Just wish I had properly tested the fans before mounting them...

Thanks for the info guys.
 

·
Registered
Joined
·
2,055 Posts
Hi there

I have only tried or run 6*EK Vardar F3 1850RPM on my loop and X99 ASRock X99 Extreme6 and Aquaero 6 XT and must say on this build no issues with fans ramping up during the boot up or during the cold boot, never run to issues with them, only one thing I hated on these fans which have been whining noise in some specific RPM and overall loudness of fans

On other build where friend has run Asus Rampage V Extreme and there we are run Aquaero 6 Pro with similar fan setup and radiator setup(360mm on top and on bottom 240mm 60mm thick radiator), on top we are run Phanteks PH-F120MP and on bottom 240mm 60mm thick radiator we are run EK Vardar F3 1850RPM which would ramp up during boot up to max RPM and then during Windows logo fans would return to normal PWM settings, sometimes fans wouldn't slow down to normal PWM settings unless you are try again set them in Aquasuite to PWM

Hard to say why these fans responded so differently on Asus Rampage V Extreme and ASRock X99 Extreme6

Hope this helps

Thanks, Jura
 

·
Vermin Supreme 2020
Joined
·
25,769 Posts
isn't the ramping on boot an inherent system thing, due to getting power before settings click on?

also - doesn't aquaero include a delay start option?
 

·
EKWB Forum Rep
Joined
·
295 Posts
I feel like my fans have always span at max speed briefly on boot and always thought it was related to the motherboard or inherent system designs. It's not something I'm very knowledgeable about, but I'll do my best to help find a solution if you haven't found one yet.
 

·
Master of Black Snow
Joined
·
19,338 Posts
I feel like my fans have always span at max speed briefly on boot and always thought it was related to the motherboard or inherent system designs. It's not something I'm very knowledgeable about, but I'll do my best to help find a solution if you haven't found one yet.
It's part of the POST operation. Nothing wrong with the Vardars based on that logic. :thumb:

~Ceadder :drink:
 

·
Registered
Joined
·
243 Posts
Discussion Starter #12
isn't the ramping on boot an inherent system thing, due to getting power before settings click on?

also - doesn't aquaero include a delay start option?
In this case, no. As I stated above, I tested the behavior after boot, by disconnecting and reconnecting a fan from the splitter. I ramps up to full speed and then comes back down to the PWM setting.
Aquaero includes a start boost function, but because it is working in PWM and the Vardar ignores this signal until after the full spinup, it doesn't work. It would have been helpful if the boost function could be set as voltage controlled but keep the fans after that as PWM controlled.

I feel like my fans have always span at max speed briefly on boot and always thought it was related to the motherboard or inherent system designs. It's not something I'm very knowledgeable about, but I'll do my best to help find a solution if you haven't found one yet.
Thank you for the support. It can be easily tested with excluding the underlying system by disconnecting and reconnecting the fan after boot.
 

·
Master of Black Snow
Joined
·
19,338 Posts
In this case, no. As I stated above, I tested the behavior after boot, by disconnecting and reconnecting a fan from the splitter. I ramps up to full speed and then comes back down to the PWM setting.
Aquaero includes a start boost function, but because it is working in PWM and the Vardar ignores this signal until after the full spinup, it doesn't work. It would have been helpful if the boost function could be set as voltage controlled but keep the fans after that as PWM controlled.-------- Thank you for the support. It can be easily tested with excluding the underlying system by disconnecting and reconnecting the fan after boot.
Fans are NOT hotswappable. Why would you do that? You're pushing electricty through everything on the MB. Pulling fans or any component(for that matter) from a motherboard could lead to issues that bork the connection you unplug from at minimum. At worst you could not only fry your board, but bork the component you unplugged from it. Doing this kind of thing makes little to no sense. :blinksmil

And YES that is exactly what is SUPPOSED to happen during POST. Your system is making sure that everything is in working order going through the POST process. Power On Self Test exists for a reason. :mellowsmi

~Ceadder :drink:
 

·
Registered
Joined
·
243 Posts
Discussion Starter #14
Fans are NOT hotswappable. Why would you do that? You're pushing electricty through everything on the MB. Pulling fans or any component(for that matter) from a motherboard could lead to issues that bork the connection you unplug from at minimum. At worst you could not only fry your board, but bork the component you unplugged from it. Doing this kind of thing makes little to no sense. :blinksmil

And YES that is exactly what is SUPPOSED to happen during POST. Your system is making sure that everything is in working order going through the POST process. Power On Self Test exists for a reason. :mellowsmi

~Ceadder :drink:
Never have I seen such utter nonsense.

First of all I was speaking about Aquaero and not the MB,
Second, reconnecting a FAN to a FAN header, be it on the MB or Aquaero does nothing catastrophic to the source. The warnings from all manufacturers are to mitigate short circuits and other problems from misaligned connection.
Third, it was purely for testing and thus coming to the conclusion that indeed the EVO Vardars ignore PWM signal during spinup
Fourth, during POST the MB checks and alerts you about fan speeds IF you have set up an alarm for the specific fan header.

Anyway, I have tested the gray Vardars and they do not behave as the EVOs. I believe that the EVO do this in order to achieve the low RPM on any platform.
 

·
Master of Black Snow
Joined
·
19,338 Posts
Never have I seen such utter nonsense.

First of all I was speaking about Aquaero and not the MB,
Second, reconnecting a FAN to a FAN header, be it on the MB or Aquaero does nothing catastrophic to the source. The warnings from all manufacturers are to mitigate short circuits and other problems from misaligned connection.
Third, it was purely for testing and thus coming to the conclusion that indeed the EVO Vardars ignore PWM signal during spinup
Fourth, during POST the MB checks and alerts you about fan speeds IF you have set up an alarm for the specific fan header.

Anyway, I have tested the gray Vardars and they do not behave as the EVOs. I believe that the EVO do this in order to achieve the low RPM on any platform.
Don't say you weren't warned. I realize I was looking at this from a MB only perspective and not a controller perspective. But it's not an advisable process you use to get the outcome you did. Are you connected to the CPU through a PWM only header? I have hubs that have the feature and unless that feature is used, any fan connected in the hub will run @ 100percent.

I have the 500-2200 ER EVO fans and they all ramp up to 100percent during POST. They then go back to board setting. I believe my low setting is @ 750rpm. So yes unless you got a locked speed variant, I seriously doubt that the PWM feature doesn't work. I mean yes the fans could be bad out of the box, but I have a hard time believing there is a performance discrepancy between the Gray Blade EVO and the ER EVO. :thinking:

That said, if you were to pull a component from your MB while it's powered, it can cause shorting of the connection if you do so. I would never advise anyone to do that. And where my concern arose. :mellowsmi

~Ceadder :drink:
 

·
Registered
Joined
·
243 Posts
Discussion Starter #16 (Edited)
To keep it simple since it seems that we got lost in translation somehow.

I noticed that the Vardar ER line from 2017 were always starting at 100%. At first I thought that it was the MB (the Asus boards do this) but then I moved them to AQ. Unfortunately they behaved the same.

To be able to see if it was the fan itself doing this and not the underlying controller, I disconnected and reconnected the fan while the system was working and the controller was sending the PWM signal that I configured. The Varder ER did the same thing. (the gray Vardar behave as expected, they follow the PWM setting from the beginning)

I then asked the forum if the new line of Vardar EVOs behave in a similar fashion and it seems they do. That is: they ignore the PWM signal for the first few seconds after they receive power and only then they conform to the PWM setting. They do this so they can achieve around 400RPM minimum speed. It's basically an integrated start-boost.

This behaviour, unfortunately, poses some problems when you want to use a capable controller like the AQ:
- you cannot have a passively cooled system when not under load (by using a set point controller without holding minimum speed)
- if you try it, every time the controller sends other PWM signals > 0%, the fans will spin up at full speed thus defeating the purpose of a silent PC (having 16 fans all ramp up is not a beautiful symphony)
- you stress the fan controller outputs unnecessarily. 6 x Vardar at spin up go over 3Amps. If they would have behaved as a standard PWM fan, you could have doubled that number on a single header without any issues.
- the only way to achieve a slow ramp up is to define the fan header as voltage controlled. This way the fans won't have the power to do their start up routine.

As a side note: I've hot swapped fans a reasonable amount of times due to different reasons and I have never encountered a problem (most server platforms use hot swappable fans and the only difference is that they have a slot and guide the fan). 2 things you should mind: properly align the header and plug before connection and plug it quick. It is not the same thing as hotswapping a PCI card, RAM, etc.

If you are worried that something might break doing this, but still want to test the Vardar EVO behaviour, use your MB fan control software and turn off then on the header and see what the Vardar does (under BIOS or OS, doesn't matter)
 

·
Master of Black Snow
Joined
·
19,338 Posts
To keep it simple since it seems that we got lost in translation somehow.

I noticed that the Vardar ER line from 2017 were always starting at 100%. At first I thought that it was the MB (the Asus boards do this) but then I moved them to AQ. Unfortunately they behaved the same.

To be able to see if it was the fan itself doing this and not the underlying controller, I disconnected and reconnected the fan while the system was working and the controller was sending the PWM signal that I configured. The Varder ER did the same thing. (the gray Vardar behave as expected, they follow the PWM setting from the beginning)

I then asked the forum if the new line of Vardar EVOs behave in a similar fashion and it seems they do. That is: they ignore the PWM signal for the first few seconds after they receive power and only then they conform to the PWM setting. They do this so they can achieve around 400RPM minimum speed. It's basically an integrated start-boost.

This behaviour, unfortunately, poses some problems when you want to use a capable controller like the AQ:
- you cannot have a passively cooled system when not under load (by using a set point controller without holding minimum speed)
- if you try it, every time the controller sends other PWM signals > 0%, the fans will spin up at full speed thus defeating the purpose of a silent PC (having 16 fans all ramp up is not a beautiful symphony)
- you stress the fan controller outputs unnecessarily. 6 x Vardar at spin up go over 3Amps. If they would have behaved as a standard PWM fan, you could have doubled that number on a single header without any issues.
- the only way to achieve a slow ramp up is to define the fan header as voltage controlled. This way the fans won't have the power to do their start up routine.

As a side note: I've hot swapped fans a reasonable amount of times due to different reasons and I have never encountered a problem (most server platforms use hot swappable fans and the only difference is that they have a slot and guide the fan). 2 things you should mind: properly align the header and plug before connection and plug it quick. It is not the same thing as hotswapping a PCI card, RAM, etc.
Ahhh gotcha. I think I've covered your method enough so I won't continue to beat the dead horse. And I do agree that it's not the same as doing so with PCI(e), RAM or SATA connections. You already know I wouldn't suggest it. :)

As far as the cacophony of fans upon startup, I really don't consider 15 seconds of Vardar Jazz that big of an issue. If nothing else, I simply leave my system on 24/7 to avoid the issue altogether. But I get it. Some Enthusiasts like their peace and quiet as well as their sanity. :D

If you are worried that something might break doing this, but still want to test the Vardar EVO behaviour, use your MB fan control software and turn off then on the header and see what the Vardar does (under BIOS or OS, doesn't matter)
I would recommend doing it this way every time. Sure saves unnecessary wear and tear on the peripheral as well as the board. :thumb:

~Ceadder :drink:
 

·
Registered
Joined
·
243 Posts
Discussion Starter #18
To me, it's not even that (noise related), but after two years of having this setup, my AQ6 current sensor is basically worn out. It cannot sustain 6 fans on the header while they spin up to full speed because of high current.

I'm the beginning it was fine with 10 per channel, then after a while with 8 and then with 6. Now 5. After the spin up boost, 6 fans only pull 1.4 A at full speed.
 

·
Master of Black Snow
Joined
·
19,338 Posts
To me, it's not even that (noise related), but after two years of having this setup, my AQ6 current sensor is basically worn out. It cannot sustain 6 fans on the header while they spin up to full speed because of high current.

I'm the beginning it was fine with 10 per channel, then after a while with 8 and then with 6. Now 5. After the spin up boost, 6 fans only pull 1.4 A at full speed.
I think that's your answer right there maybe?

Like that maybe over time your AQ6 has lost electrical integrity? Are you running or have you tried to run 10 of the original Vards in place of the ER? Is your system set up on an UPS or is it on a surge protection device? If it's not on either I would be surprised but not beyond the realm of possibility. If not then it could be the reason you are seeing such a degradation over one channel. Is this just the one channel or does it happen on the others? :mellowsmi

Cause @ 21.6w/10 ERs' that indicates the possibility that that channel is limited or maybe you're controller is on its way out. Now way your 30wpc cabaple connection should only be able to handle ~11w of power. Maybe instead of looking for an answer with the fan, maybe you're better served with diagnosing the issue with the controller first? I could be wrong but I doubt the fan or number of fans is the issue short of the number that can reside on one 30w channel. ::thinking:

~Ceadder :drink:
 

·
Registered
Joined
·
243 Posts
Discussion Starter #20
It's not the fan header or electrical capability of the controller itself, but the current monitoring chip (mine errors out while the Vardars "boot up" even on channels that do not have load, only pump PWM signal). I've confirmed this with Aqua Computer support and they've seen issues with Vardars that behave this way.

From an EE perspective I do not see why having an UPS (unless its a Double Conversion Online type - which aren't for home use) has anything to do with this topic. A typical Line Interactive UPS has 1-2 steps buck-boost and doesn't do much when power fluctuates (it also goes through 0V and relies on the PSU to hold up power until it switches to battery). I think you are stuck on electrical problems generated by me disconnecting and reconnecting the fans :)

The reason I opened this topic was to know if the EVOs behave the same way in case I decide to replace my fans.
 
1 - 20 of 38 Posts
Top