Should this fan be this noisy?

[Randall]

Marshall MG100DFX in with the power chip blown up and the fan missing. Replaced the TDA7293 power module, and fan with with what I think is a suitable 12v replacement. But it sounds like an angry bee. It is oriented to blow out as an exhaust fan. The strange thing is there are no vents for air flow in the chassis at all, only the 1/4" jacks. So this fan is strapped to the heatsink pulling air past the fins to the outside.

It is quite loud for a fan. Curious there are no air flow vents anywhere. I don't have anything to reference it by, as I have not heard one of these before. Do I have a crap fan, or is this all that can be expected for this arrangement?

[Jon Snell]

According to the Specification Sheet on this fan, the noise level is 37db. Fairly quiet as fans go.
If louder than that, I would suspect particle build up on the vanes or a noisy/worn bearing. Check it is fitted correctly and not catching on the heatsink.
These are brushless low voltage DC fans.

[Mick Bailey]

I've replaced a few of these fans due to noise and sometimes a new fan has been nearly as loud as the old one, so I've had to audition different fans to get one that's acceptably quiet. The odd thing is that when removed the noisy (new) fan is fairly quiet just on the bench. Maybe the constricted air flow places a different dynamic on the fan blades or bearings.

[Helmholtz]

Any fan will be louder when built in.
In free air it hasn't much work to do (low air resistance).
Quite the opposite when it has to fight against a closed air volume. Think a vacuum cleaner with a clogged nozzle.
The other reason for increased noise is that it makes the chassis vibrate when mounted.

You might try to lower the voltage with a series resistor, e.g. 50R/2W.

[Randall]

This is a new fan, so it isn't dirty or restricted. It is installed correctly I THINK, with the label facing out and drawing air out. I'd rather not try to re-engineer it by playing with voltages, no money in the budget for that. If this is how it was designed and sold, than it is what it is, I guess.

[eschertron]

Marshall MG100DFX in with the power chip blown up and the fan missing

Did the current customer remove the fan? Maybe because of the noise? If so, you have reason to be concerned, otherwise...

This is a new fan, so it isn't dirty or restricted. It is installed correctly I THINK, with the label facing out and drawing air out. I'd rather not try to re-engineer it by playing with voltages, no money in the budget for that. If this is how it was designed and sold, than it is what it is, I guess.

Then it is a proper repair.
If the customer wants to provide you with a higher-quality noiseless fan, let them. Or let them punch holes in the enclosure. It's their equipment.

[Randall]

There was some of the old fan connected to the heat sink when I got it. It was a thrift shop basket case. Suspect the fan died, and that cooked the power module.

I reversed the new fan to blow air into the heat sink. Problem solved, it is quiet enough now.

[g1]

Curious there are no air flow vents anywhere.

Doesn't the chassis have open sides on these? I think they rely on the gap between the chassis and the cabinet.

[TomCarlos]

I worked on one Marshall MG series amp... the MG250dfx. That amp has 1 fan that cools a much larger Heatsink-PC board that has 2 power amp modules (because it is a two channel amp). So the cooling was horrific. I thought about removing the Heatsink-PC board assembly and mounting it in the case, somewhere "near" the fan opening. That way, the fan would not be obstructed when pushing air into the case and across the heatsink. It is just an idea.

Take a look at the recommended replacement fan at Full Compass. Remember, the greater the air flow, the greater the noise.

https://www.fullcompass.com/prod/514769-marshall-m-fans-90002-fan-for-mg50dfx-and-mg100dfx

[tedmich]

if you can tolerate a slightly (15mm) thicker fan (the heatsink/PCB free floating?) you could go with a Delta AUC0512DB-AF00 for about $10 too.
It moves more air at 20 CFM and with some o-rings its probably as quiet. (McMaster Carr 2418T113 or equivalent , Dash 007)

For $1.60 theres a nice silicone gasket (Digikey RPC3978-ND) or for the ultimate dampening you can cut down some McMaster Carr sorbothane (3462K11)

[Randall]

Per post #7, reversing the fan air flow corrected the objectionable noise. Problem solved.

[J M Fahey]

Chiming in to suggest blowing air INTO the heatsink is the way to go, glad you found that on your own.
Way better cooling because air is way more turbulent being just beaten by blades than leisurely sucking .

I guess IBM chose the opposite to minimize dust/lint collection and of course everybody else just followed blindly.