M-Cubed miniNG......mini review.

Control: management of fans, temp/rpm monitoring via soft/hardware

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

M-Cubed miniNG......mini review.

Post by Bluefront » Sun May 21, 2006 3:12 am

The M-Cubed miniNG fan controller has been around for a while, but I have been able to find little info on the device, and the manual is confusing to me. Now I know why.....it's a complicated device for being a fan controller. My impressions of the thing (I bought two) are based on three days of use, attempting to get it working with my single-fan Aria (in the Gallery section). This computer uses one fan, a Globe S1202512L-3M. It has it's own temp sensor shorted, so it acts like a normal fan, with a max of 1800rpms. Modded like this it starts at 5V every time. I briefly tried the miniNG with a Yate Loon fan.....the operation was the same with both fans. Since my computer setup is somewhat unusual, your own experience with a miniNG could be different from mine. Here's a photo of my setup, the miniNG mounted externally, under a hinged pod (makes adjustments easy).

Image

Image

I'll start off with the negatives.....might save you some reading.

This device disables your rpm monitoring ability. If you try to read the yellow rpm wire the miniNG goes into a fail-safe mode, and the fan goes to the max.

The adjustment pots are recessed, turned with a supplied screwdriver, and very sensitive. It's impossible to return to a previous setting without the use of a voltmeter (not supplied).

When you make pot adjustments while the computer is running, the miniNG is slow to respond. This means you have to make very small adjustments and wait for a minute or so till the thing responds. This makes setup a time-consuming procedure.

This controler apparently will only work with three-wire fans. It needs to sense rpm for correct operation.

The eight jumpers are also recessed, very difficult to deal with. once installed they are hard to remove.

But I guess the worst thing for me was the confusing setup, with a small manual with little directions for use.

Ok....on to the device. It has five different operation modes, including one mode that interfaces with a T-Balancer(I'm using mine in a stand-alone mode).....and there are two completely separate channels for two fans, with a different sensor wire for each fan. Each channel can be setup differently. Plus there are a few other settings. You can run either analog voltage or PWM. The PWM will handle 2x the current, enough to run anything I can think you would do with this device. At low voltage using PWM, my Globe fan makes clicking noises....so I'm using analog. There is also a fail-safe mode for fan start-up. This controller gives a blast of 12V for three seconds at the computer start-up. But it also can turn off the fan during operation. If the fail-safe mode is enabled, the fan voltage is gradually increased till the fan starts spinning, then the voltage drops back.

Manual mode......you can set the fan voltage using the pot. Really low. You get the 12V boost to start spinning, and can go down to about 4V if your fan will run that low. Apparently there is a high-temp over-ride in all the modes.....over about 70C the fan goes to the max.

Step mode.....you set an idle voltage, and the fan goes to the max when a adjustable temp is reached.

T-Balancer interface mode.....there is a connection wire to run to a T-Balancer. This gives you additional abilities and adjustments. Since I don't have a T-Balancer, I cannot comment on this mode. From reading the manual, it appears this mode might work better for me, at further complication and expense.

Target Temp mode.....you set a temperature using the pot to adjust voltage, then the fan speeds up/down as the temp changes. I tried this mode. I set a temp of 45C running CPUBurn (about 7V with this Aria setup). That works ok, but when I turn off CPUBurn, and the computer idles, after a minute or so the fan slows down till it gets 0V and stops. Then the CPU heats up and the fan starts going again when the temp gets close to 45C. In my setup with only one fan this mode is not useable...since everything else heats up without any airflow. In other setups this mode might be ok.

Curve mode.......this is the mode I'm using. It was tough to get the curve adjusted. Right now my setup idles about 4.5V @ 33C (CPU temp at 25C ambient). Using CPUBurn the voltage gradually rises to a little over 7V at 45C. The voltage rise is gradual, no steps. It gives a sound increase not noticeable. I like it. I could adjust the idle voltage down to 4V, but then the max temp goes up to close to 50C. This is where the T-Balancer mode would help out, as I could modify the curve. Right now I can move the temp/voltage mode lower/higher, but the curve stays the same. (I know it's hard to explain).

Overall I'm keeping the miniNG. It's a good deal for the $40 I paid, and provides many options. If you don't mind a complicated setup procedure, this thing might be the fan controller for you. There are other aspects of this device I haven't touched on....maybe I can answer other questions if this review leaves you in the dark...... :lol:

Few more things.....Using only one fan, the device stays completely cool, requiring no airflow. The placement of the temp sensor is very important if you are using the miniNG to control a CPUfan. On the underside of the heatsink close to the edge of the CPU is the best place. It comes with some velcro and tie-wraps to mount it, along with two sensor wires, a wire to connect to a T-Balancer, and a screwdriver to adjust the pots. It took 10 days normal shipping to get the thing (Austria to ST Louis USA).
Last edited by Bluefront on Sun May 21, 2006 5:21 am, edited 4 times in total.

sanse
Posts: 399
Joined: Thu Dec 22, 2005 9:48 am
Location: Netherlands

Re: M-Cubed miniNG......mini review.

Post by sanse » Sun May 21, 2006 3:50 am

Bluefront wrote:My impressions of the thing (I bought two) are based on three days of use, attempting to get it working with my single-fan Aria (in the Gallery section).
hi,

i just extended my t-balancer sl4 setup with one of these mini-ng controllers. mainly because 2 fans in my rig still were a bit (pwm) noisy even after putting in attentuators in between.

the fans are dead-silent now in analogue-mode, but i also noticed the rpm-problem. it's not simply possible to extend the yellow and black wire back to the mobo to get a rpm-reading there.

apparentely the mini-ng is a 'negative'-side fan-controller; connecting the black fan-wire back to the mobo effectively bypasses the mini-ng and indeed the fan starts running then at full speed.

i anyone knows a solution to this rpm-problem i'll be obliged.

for the rest the device is very neat, when using/configuring it from within the t-ban software. it allows then to transfer rpm-temp-curves for saving in the mini-ng itself. so if you configure it with jumpers to run in curve mode you don't need the poti's; it just boots itself with the curve you transferred to it with the t-ban software.

btw, i read there will be a new 4-channel t-balancer soon with this same dual-mode analog/pwm technique.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sun May 21, 2006 5:31 am

I spent a bunch of time trying to read rpms. Even if you only attach the yellow wire to the board, the miniNG hangs up. I tried using an external device (the second photo) to read rpms....still doesn't work, and the miniNG stops responding. Lost cause I think.... :(

sanse
Posts: 399
Joined: Thu Dec 22, 2005 9:48 am
Location: Netherlands

Post by sanse » Sun May 21, 2006 5:40 am

Bluefront wrote:I spent a bunch of time trying to read rpms. Even if you only attach the yellow wire to the board, the miniNG hangs up. I tried using an external device (the second photo) to read rpms....still doesn't work, and the miniNG stops responding. Lost cause I think.... :(
i think the mini-NG does not hang up; it just gets by-passed by the black-wire to the mobo. the result is a full-speed fan with mobo-rpm-reading.

if you only connect the yellow-wire the mobo has no reference and the mini-NG rpm-circuit gets confused by the pullup-resistor on the mobo. the result then is a mini-NG controlled fan with no rpm-readings at all; nor the mini-NG nor the mobo.

if anyone has a simple solution to this rpm-problem?

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sun May 21, 2006 12:43 pm

Here's a temp graph I just made showing the curve mode at work. At the point I started two instances of CPUBurn, you see a smooth, gradual temperature rise, peaking at 46C. There are no peaks or steps in the curve...the voltage raises gradually along with the temp, and ends up about 7.6V. This really makes the slightly increasing fan noise level, almost un-noticeable unless you really consentrate. Works well in this mode....

Image

jmke
*Lifetime Patron*
Posts: 495
Joined: Thu Apr 15, 2004 3:53 am
Location: In Front of PC
Contact:

Post by jmke » Wed May 24, 2006 2:25 am

Hello Bluefront, excellent mini (maxi?) review of this fan controller, I got one miniNG in for review from mCubed, will hook it up to a T-Balancer also to see how I can change the fan speed/fan curve.

It seems that the plethora of options makes it a tad user unfriendly...

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Wed May 24, 2006 3:45 am

Yeah that's the strong point and the weak point of this device. Many adjustment options at increased complication. Can't have everything I guess.

Please report back on how this thing works with a T-Balancer. I'm very interested. :)

jmke
*Lifetime Patron*
Posts: 495
Joined: Thu Apr 15, 2004 3:53 am
Location: In Front of PC
Contact:

Post by jmke » Mon May 29, 2006 2:50 am

this mini controller is feature packed, almost too much for its price:) the fan control is very expanded, the best config in standalone is without the doubt the corrected Curve, where the alarm temp is 70°C and you alter the min. fan speed using the small dial. the fan will speed up and slow down gradually and wont cause any erratic noise increases. works great this way.

after hooking it up the T-Balancer you can configure the curve the way you want it, setting a lower max. alarm temp, you can also configure the fan blockage alert as well as the PWM/Analoge power setting from within the software config panels of the T-Balancer.

The latest software version of the T-Balancer adds some very useful features, it's almost 2 years ago I tested the T-Balancer, the new firmware update and control center software has changed quite a bit. The best addition is the possibility to use MBM5 and Speedfan readouts to set configure the fan's speed up curve, so you don't have to rely on those temp probes alone anymore, you can configure both and if one fails the other will take over, very nice.

You can now control the CPU fan through the T-Balancer by reading out the values from Speedfan.

All in all, even without the T-Balancer, the miniNG impressed me, the PWM output allows you to let the fan very slowly, but it doesn't work well with all fans, using it on some will actually make the fan noisier (the fan motor acting up). With Analoge power I could get some fans to work with only 3-4v, the anti-block detection will increase the fan voltage incase the fan does stop working. works quite well.

I'm currently finishing an AM2 platform review (3500+ S939/AM2) but miniNG review is to follow that one just thereafter; but from what I've written down so far, not many drawbacks, except for the price of course, which is not cheap for a fan controller, even if it's the most advanced little thingie out there, if you look at in from an end-user point of view, that €40 might be better spent on some other aspects for quiet computing? It will come down to personal preference... like always.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Mon May 29, 2006 1:18 pm

To me it makes sense not to compromize on your fan control. After all, the fans are the noisiest things in most computers, but with a good control system the noise can be toned down. Very few boards I have used/owned had a good fan control.

This miniNG is the answer for me....

MalcolmC
Posts: 67
Joined: Thu Mar 17, 2005 2:05 pm
Location: UK

Post by MalcolmC » Tue May 30, 2006 1:58 am

I posted few comments about this in this section on 8 May, and since then my views have not changed. The instructions are poor, but the product itself is fantastic value for money, and does just about everything you could want from a fan controller.

Personally, I am more than happy with the product, I've got mine attached to a T-balancer, and I do not find the software intuitive, but once you've spent a bit of time, you get used to it.

dinofx
Posts: 36
Joined: Thu May 25, 2006 12:36 pm

Post by dinofx » Wed Jun 07, 2006 11:54 am

Perhaps this device is using an N-channel MOSFET instead of the less popular P-channel. This would mean that in analog mode, the fan's ground is reaised instead of its postive voltage lowered. This would explain why the tachometer fails, as the fan would no longer have a true ground.

Can one of the owners of this circuit read the part number on the MOSFET? It's the large-cased part with a metal tab sticking out of it mounted to the heatsink.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Wed Jun 07, 2006 12:49 pm

I don't think that part is visible.....it must be between the board and the heatsink. I'd have to take the thing apart. Don't really want to do that while I have a warranty.... :(

dinofx
Posts: 36
Joined: Thu May 25, 2006 12:36 pm

Post by dinofx » Thu Jun 08, 2006 6:41 pm

It's most likely that one screw in the middle to remove the heatsink, but even so you might be looking at the back side of the MOSFET. A voltmeter would also answer the question. Just compare ground to the fans ground at less than 100% "throttle"

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Fri Jun 09, 2006 2:10 am

There appears to be some TIM between the board and the heatsink....I really don't want to disturb anything as the thing is working right now. :lol:

Get one yourself.....it's a neat little device that might give you some ideas. I've already caused the thing some grief by trying to read the rpm wire. I expect to use this second unit shortly. So I don't want to mess it up.

Post Reply