Rigol DS1054Z Fan Replacement

[andersjm]

Hello forum!

I thought I'd share my experience in replacing the fan on the Rigol DS1054Z. I know there are other threads about fan replacement, but not one dedicated to the DS1054Z, I think.

Anyway, having good experience with "Pabst" fans previously, I went and bought an EBM/Pabst 512F 2. I wasn't sure it was going to fit, I just knew that the outer dimensions were similar to the stock Sunon. It turned out, it didn't. Well, it didn't have an identical screw pattern, but it was possible to fit it anyway.

Link on digikey: http://www.digikey.se/product-detail/en/512F%2F2/381-2682-ND/5043858

I used the built-in acoustic noise measurement capability of my Android phone.
I put a couple of marks on my kitchen table for where to put the DS1054Z and the phone, so that I could do precisely similar before- and after-measurements.

With scope off: 27dBa
Before(stock fan): 43dBa
After(Pabst): 39dBa

The Pabst fan has a higher nominal flow, and is rated at 1.0W versus 0.78W for the stock Sunon. Hopefully, the DS1054Z can handle the extra current draw.

The Pabst fan has the holes slightly farther apart from each other (1-2 mm). I used slender M2.5 machine screws with nuts, instead of the thick self tapping ones, and the thinner screw meant I could mount the screws slightly off-center, providing a reasonably good mounting.

Considering the small difference in noise level, I wouldn't say it was worth it. But it's always fun to tinker :-). So while the Pabst can be made to work, I'd say there are probably better options.

[pascal_sweden]

The difference in noise level is almost nothing...

The fan did not fit in well... and it draws more current...

Would have made more sense if you would have done some research first on different fans, to find a fan that is a drop-in replacement and which has a much lower noise level.

Now you have lost your warranty on your scope, and it probably does not look so nice anymore either, and in the long run might even damage your scope.. all this for a lousy 4 dBa saving.

Why do you provide a link to a fan that does not fit and has no significant savings in noise level?

Do you want to encourage other people in making the same mistakes as you?

Moral of the story: don't mess around with your scope unless you first do some thorough homework and really know what you are doing.

[Augustus]

The whole scope draws about 20w from the wall, if I remember that correctly. Why on earth does that thing need a fan at all? 

[madires]

Sometimes it's not the fan but obstacles in the air flow which create most of the noise. When I look for a quiet fan I usually look for computer fans first. If you don't mind fancy modder stuff you can get some nice, quiet and inexpensive fans.

[pascal_sweden]

Maybe this is a good place to start looking for a quiet fan:
http://www.quietpc.com/casefans

[Fungus]

The whole scope draws about 20w from the wall, if I remember that correctly. Why on earth does that thing need a fan at all? 

A lot of people have asked that same question. No good answer yet...

Does it really need a fan in a normal environment? Where are the people with thermal imagers, etc., when you need them?

[SeanB]

Probably half the power is used in a single ASIC, so that either needs a massive passive cooler or a smaller finned block and a fan.

[Towger]

Sunon are good fans. They have a range of different types (of different quality) within that different speeds/cfm per type etc. You can look at their data sheets to see the differences.

[Augustus]

Sometimes it's not the fan but obstacles in the air flow which create most of noise. When I look for a quiet fan I usally look for computer fans first. If you don't mind fancy modder stuff you can get some nice, quiet and inexpensive fans.

Yeah, the fan is directly mounted to the sheet metal casing, with its airflow going straight through sharp edged cutouts. That's how they did air-raid sirens, should work on scopes also. Of course never ever use rubber grommets to dampen motor vibrations, they should be directly and effectively coupled to the device casing in order to maximize the aural experience of the proud new owner. Feel, err, hear the power

A "case modder" would just dremel that darn cutout section away and replace it with a proper wire fan guard. Any low rpm, low vibration fan replacement should do, given the low power consumption of the device. Mounted on some "Case Spätzle" of course, don't know the englisch name for them, they look like this:



Is it worth all that hassle? Probably not. If things go wrong one day you'll bite your own ass if they reject your warranty claim

[pascal_sweden]

Are noisy fans really that much cheaper than quiet fans?

Rigol is aware AFAIK that their fans are noisy, and still they are not taking any actions against it?

With their capacity in production they should be able to negotiate a good price on quiet fans, and even do small changes to the design on how the fan is mounted with rubber feet, and how the "grill" can be optimized to avoid noise from air flow passing through.

[Towger]

Once the fan is a decent brand name, which Sunon is the fan size, cfm and more importantly case/grill design are the main factors.  Most of the 'silent' pc upgrade fans push less air, so of course will be quieter.

[electr_peter]

The whole scope draws about 20w from the wall, if I remember that correctly. Why on earth does that thing need a fan at all? 

Try running at max sample rate for some time and then touch input BNC - turned on channels will be warm. 20W(?, I have not checked) means nothing in comparison if a single small part can consume few watts and overheat.

Sometimes it's not the fan but obstacles in the air flow which create most of noise.

I my experience fan on on DS1054Z is of good quality and do not emit much noise. But airflow does. Thus changing the fan will achieve no significant gain (possible losses from a bad replacement come into play). Airflow comes through multiple holes in metal/plastic cases, so no easy solution to remove sound via HW modifications exists. One can turn off the fan or temperature control it, but I would not do that on my scope. Also, there is a single fan in the unit with several significant hot parts, so how would you do temperature control?
However, even if sound "purist" disables the fan, relay clicks are still audible

On practical level, sound from DS1054Z is not annoying in pitch and is only audible in silent environments. Solution -> add some noise (open the window, add air con/fridge, turn on music or other noise sources, etc.) and you will not hear your oscilloscope.

[Augustus]

Try running at max sample rate for some time and then touch input BNC - turned on channels will be warm. 20W(?, I have not checked) means nothing in comparison if a single small part can consume few watts and overheat.

Yeah, that ~20W figure I mentioned was out of my memory, checked that on my own 1054Z a few weeks ago. But just to be sure, a quick retest:



Yup, pretty close, no Alzheimer yet, 21 watts power draw with fft numbercrunching turned on. Turning on one channel or all four doesn't make much difference either, maybe 1 watt or therabouts. If I were Rigol I would put some really big honking energy-green-tree-and-worldsaver sticker right on the front of the box

btw: aren't the frontends not cosily shielded in those silvery cans? They wouldn't get much if any airflow anyway... If the asic gets to warm, just throw a reasonably sized heatsink on that darn thing and get rid of the fan. Or at least turn it down. Turn it waaaaaaaay down... 

[mr.os]

I replaced the fan in the scope with one of this ones: http://www.blacknoise.com/site/en/products/noiseblocker-it-fans/nb-blacksilentfan-series/50x50x10mm.php?lang=EN
I'm not sure, which one, because I had them lying around.

It's less airflow, but who cares? The outgoing air is not much warmer as before and the scope runs for hours without problems. It's now much quieter and it doesn't bother me anymore.
I need other screws to attach the fan, but I think, this is not a real problem 

[madires]

btw: aren't the frontends not cosily shielded in those silvery cans? They wouldn't get much if any airflow anyway... If the asic gets to warm, just throw a reasonably sized heatsink on that darn thing and get rid of the fan. Or at least turn it down. Turn it waaaaaaaay down... 

I think a fan-less design would make the BOM US$ 5-10 more expensive. Large heatsinks are expensive and require a proper mounting. The fan also cools the PSU, DC-DC converters and some chips which might become a little bit too warm otherwise (and would need a small heatsink).

[andersjm]

Do you want to encourage other people in making the same mistakes as you?

No, that was not my intent.

I did change the fan on my scope. I was aware of the risk that I might damage the scope, but I chose to modify it anyway. You do not agree with this decision, but it was my decision. Anyway, what is done is done and there's no way to change that  .

I tried to write a balanced fact-based post, sharing my experience. I know others have changed the fan on their Rigol DS1054Z, so I thought my post could be of value to someone somewhere, even as a "don't do it this way"-kind of post.

Is there any part of my post that you feel gives a too positive description of this fan modification?

I tried to write the post I wished had existed before I did the experiment.

[tve]

Thanks to the original poster for starting this thread! Not sure why some people think that only posts about absolute successes are appropriate... I've come to the realization that I'm looking for a small handheld DSO for quick measurements because I don't like keeping my DS1054Z on all the time because of its noise. It's by far the noisiest thing around and I do get tired of the noise after a couple of hours. So rather than bu more gear I'm going to try and just fix the problem.

[moya034]

The solution to the fan noise in test equipment is to have more noise in the room!

In my lab, between the room's air cleaner, Solder smoke extractor, hot air station, and music in the background, fan noise from any test equipment is not even a concern of mine

[electr_peter]

The solution to the fan noise in test equipment is to have more noise in the room!

I agree. In a sufficiently silent room you start to hear lights flicking, clock ticking, noises from other environments, etc. which can be more stressful than little bit of random noise.

This is simply how a human ear operates - it adapts to current noise level and starts to hear sounds around this noise level. In an quiet environment it means you start hearing amplified annoying noises. The only real solution is to bring noise level up to hide these sounds under the "noise floor". This can be done with random (pink/white/...) noise generator or some other means.