As mentioned above, I have now got the MiniDSP up and running, and I was able to use REW to generate the filters to import into the MiniDSP. I made one custom addition to boost a null, and things are sounding pretty great.
One thing, however, that I did still need to fix was the fan noise from the iNuke amp. It was enough that we generally didn't turn the new subs on for general watching, and unless we were putting on a blockbuster type movie. Even then, the fans were audible and annoying during quiet passages.
So after reading up on a lot of different people's approaches, I settled on just replacing the two 80mm fans with Noctua fans, which were said to be near silent, and most people had no issues with heat after the swap. I had considered using 92mm or 120mm fans with an adapter to get more air flow at less RPMs, but in the end decided to just go for the simplest option that was largely a proven success.
Here's how we start. I used two Noctua NF-A8 (FLX) fans, which were about $20 each from Amazon.ca. Yes, that's expensive for an 80mm fan, but Noctua's reputation for both quality and quietness is about the best there is, so I figured they were worth it. When the fans arrived, the packaging and included accessories was very impressive, and the fans are very well designed and made, unlike a $5.00 PC store special.
The first step is to remove the top of the amp case, which is done by removing six small Phillips screws. The fans also have obvious screws on the back of the amp, which also hold the grilles on (also Phillips). I have indicated where the fan headers attach to the board in this pic.
The header plugs have a dab of glue on them to hold them in place, but that wasn't too hard to pick off and loosen up enough to remove the plugs. The biggest hurdle I encountered here was the big blob of glue that was used to secure the fan wires to the bottom of the amp chassis. Oh, and of course the wires were twisted and crossed in the pile of glue, so removing them took a while. I tried to pick and pull them out, but it just wasn't working, so I heated up the glue gun and used the tip as a prod to loosen the wires and separate them. Once that was done, the rest was smooth sailing.
I replaced the two fans, which were a perfect fit, and the relatively short lead wires were just enough to run to the fan connectors on the board (circled). If I needed more length, the Noctuas came with an extension plug that would have worked fine.
I did what I could to keep the wires out of the way, and plugged them in to the board for a quick power test. Make sure you have + to + and - to - first, with the yellow PWM wire overhanging and remaining disconnected. Everything worked, so I unplugged the amp and got ready to close up. I decided that since the lead wires were JUST enough to get to the connectors, and the pins from the board were a little shorter than the plugs ideally would have taken, I wanted to make sure they stayed in place so I took a page from the manufacturer and added a little blob of hot glue to secure the two in place.
With the top back in place, I fired up the amp for another test, and had no problems. A few things of note to me after the swap:
- The fans really are a night and day difference. The Noctuas are near silent from about 2' away, and when I got the amp back in place, the hum of the LED ceiling bulb was louder than the amp.
- I am a little concerned about the volume of air being moved though. That was also a big change, and I can barely feel much air at all coming out the front of the amp when it is on anymore.
- Since many people have used these fans without problems, I'm not going to worry much, but it was definitely a significant decrease in air flow. I'd be OK with a little more noise and a little more air movement, but if this works well then it really is ideal.
- I have considered drilling out a few holes in the front of the amp faceplate, since the existing vents aren't exactly optimized for good flow. We'll see how things go.
I left the amp idling for 30 mins or so and didn't notice any issues. I only had time to run one quiet passage to test with any kind of load, and it worked fine as well. I guess time will tell, and how hard I'm driving the UXLs will be the biggest factor in the long run, but overall my first impression is that this was a success. Thanks to all who helped out with information and by posting their own procedure and results.
Oh, and just a note... I've read others say that they believe their fans may ramp up speed as heat (or load?) is detected in the amp, but I certainly haven't noticed this, and I don't know if this is even possible with the connections on the board. I guess it may be able to vary the voltage to the headers, but that seems advanced for this type of no-feature amp. I might be wrong, but that's my impression so far.
