EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Datman on April 10, 2015, 08:42:30 am
-
Hi
I'd like to talk here about the noise from the fan. The problem have been debated in a long topic, but I'd like to put together here all experiences about it.
Two days ago I had to work using the scope. It was after dinner. I was listening to the radio. When I turned on the scope I have been very hurted by the fan noise... :-(
I noticed the noise before, but I thought I could bear it. Instead, in the silence of the night it is too loud!
What are your experiences? Do you already reaplaced it and now it is quieter? It seems to be a 5cm fan, with 4cm between the screws. Is it a 12V fan?
This evening I'll go to an electronics shop near home. I'll ask there for it. Do you thimk a computer shop would be a better place?
Thanks!
-
A couple of potential replacement fans have been suggested here: https://www.eevblog.com/forum/testgear/new-rigol-ds1054z-oscilloscope/150/ (https://www.eevblog.com/forum/testgear/new-rigol-ds1054z-oscilloscope/150/)
I did buy the recommended Gelid Silent 5, but found that -- while it was indeed much quieter than the factory-installed fan -- it pushed significantly less air. I did not dare to leave it installed without measuring the temperatures of a few key components before and after; and was too lazy to do these measurements. Hence, I went back to the factory fan.
Most of the noise seems to come from the actual airstream, not the fan bearings. Hence the "silent" fans will probably all imply a compromise in air flow. You might as well install a dropping diode or resistor in the supply cable for the factory fan, if you want to accept the potential risk of reduced air flow and higher internal temperature. Installing a much larger and slower-spinning fan would be the best solution, but I don't see how it could be fitted in.
-
Yes, I also think that mostly of the noise comes from air flowing. I should remove the fan and listen to his noise in the free air.
I could install a bigger fan keeping it a few centimeters far from the grille. Being bigger, a little short circuit for the air should not be an issue. It is not very hot inside the case, but air flowing is important for cooling electronics on the board, removing heath from them.
-
Just notice about fan replacement.
I don't know how is on DS1000Z but on other Rigol devices lkie DP832, you have to check power consumption of the new fun. It should be the same as original. There is current monitoring to determinate fun speed/condition. If the current is too low device cannot turn on. Maybe it's only on power supply but don't be badly surprised.
https://www.youtube.com/watch?v=b0kTXX8zXSo (https://www.youtube.com/watch?v=b0kTXX8zXSo)
-
2 years ago, I put a resistor in series with the fan on my DS1204B.
With a potentiometer I discovered the minumum startup value, then put a fixed resistor.
Flow is less now, I monitored the ambient temperature inside the scope, stays the same.
Nearly impossible to hear now, don't know if measurments are influenced. Everything works fine.
-
I replaced the fan in my DS1054Z months ago with this one:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835352002&nm_mc=TEMC-RMA-Approvel&cm_mmc=TEMC-RMA-Approvel-_-Content-_-text-_- (http://www.newegg.com/Product/Product.aspx?Item=N82E16835352002&nm_mc=TEMC-RMA-Approvel&cm_mmc=TEMC-RMA-Approvel-_-Content-_-text-_-)
It's much quieter. Yes, it moves less air but it still more than enough to keep the scope on all the time without overheating assuming that you're in a room temperature environment with plenty of space around the scope.
-
I could install a bigger fan keeping it a few centimeters far from the grille. Being bigger, a little short circuit for the air should not be an issue. It is not very hot inside the case, but air flowing is important for cooling electronics on the board, removing heath from them.
I don't think that will help much. The two grilles (in the sheet metal and the plastic cover, if I remember correctly -- don't have the scope in front of me now) will continue to constrict the airflow. They will continue to create noise, and will probably cause a significant part of the fan's air flow to go into the "short circuit". Taking a saw to the grilles would probably be quite effective, but I don't think I would like the cosmetic result... ;-)
If I should get really annoyed with the fan noise of my DS1054Z at some point, I would probably just put a resistor in series with the existing fan, and measure the temperature inside the case before and after.
-
If I should get really annoyed with the fan noise of my DS1054Z at some point, I would probably just put a resistor in series with the existing fan, and measure the temperature inside the case before and after.
The stock Sunon fan, even at its impressively low starting voltage of 4V, has a somewhat annoying whine in free air without any obstructions.
I replaced it with a slower Sunon fan (http://www.reichelt.de/index.html?ARTICLE=110413 (http://www.reichelt.de/index.html?ARTICLE=110413)) which doesn't have that whine, but when installed in the DS1Z it isn't quite silent due to the case turbulence. I will probably open it back up and add a series resistor.
So to have this scope be silent, I think you will need both a replacement fan and to slow down that replacement fan.
-
I just received my 1054z a couple days ago, from the new shipment to TEquipment, and have been running it continuously since then doing various tests and evals. The fan is not nearly as noisy as I expected from reading the reviews! It's the least noisy fan that is running in my room! I've got an airconditioner, a computer, an electrostatic dust precipitator all going pretty much 24-7 and I sure can't hear the scope's fan over all that!
I even bought a Fractal Design FD-FAN-SSR2-50 fan based on the posts in the forum, but I see no need to change fans at this point! (The FD fan is nearly silent, though, when tested in open air, and seems to have good reviews and support from the company.)
Also the probes aren't nearly as bad as I expected them to be. Maybe the quality control has improved....
BUT::: I think I found a minor software error. I'm starting another thread to discuss this matter, please check it out.
FW 00.04.02.SP4
Board 0.1.1
unlocked without difficulty using code from downloaded keygen Linux command-line application from the Riglol zipfile, compiled for my system
-
I think the simplest and best solution is deflecting noise away, to the rear...
-
Dear all,
I'm as well an owner of the DS1054Z who is annoyed about the fan noise. |O.
After reading several posts, I decided to open my still "virgin" scope to look for a more quiet solution.
All solutions with other fan types are resulting in less air flow and maybe a little bit less noise - I did this as well with an NB XS1 fan.
But this was not what I expected :rant:.
So I took the long way...
First I measured the audio spectrum (!) in a few assembly states (with the rear sheet metal power supply frame solely, with both metal frames and fully assembled including the rear plastic cover).
These results were quite revealing!
They show that the metal frames are not involved in the "sound" of the scope, but the rear plastic cover, including the grille makes the difference. ::)
The peaks in the spectrum are at approx. 250...400 Hz.
By knocking on the plastic housing I found that this its resonance frequency ;D.
Calculating the fan noise (approx. 3300 rpm @ 8 volts and 7 wings) gives as well 390 Hz. So the fan noise an the housing are in resonance!
With this it is clear that the fan speed must change.
I bought a 60 mm fan (SUNON HA60151V4-000U-999, 2500 rpm @ 12 volt = 1700 rpm @ 8 volts )and tried to bring it into the housing. With some small modifications to the fan - not to the housing ;) - (space for the rivet and the chamfered edge on the rear) it fits to the metal frame.
In the first stage (as long as there is the warranty) I fixed the fan with double sided tape pads on each corner.
The result is impressive:
Very quiet and still a remarkable airflow :-+
Have fun
kaskade911
-
Here are the audio spectrum screenshots with stock equipped fan and the new 60 mm SUNON fan.
-
The result is impressive:
Very quiet and still a remarkable airflow :-+
How did you measure the airflow??
The only way to be sure you don't risk your scope is to make thermal images from the pcb with the old fan and
the new fan.
-
Hi Karel,
you are quite right, but unfortunately there is not app for airflow measurements available ;)
But both fans have nearly similar airflow figures in the data sheet.
Both are driven by the same (reduced) voltage of 8 volts.
So IMHO I supposed that the airflow is similar for both fans (and I know that the cut out in the sheet metal is designed for a 50 mm fan and not for 60 mm).
And there is still the opportunity to go up with the fan voltages to 12 volts by updating the PSU (short cut the 49.9 ohms resistor at the +FAN output)
All this solutions are better than further lowering the voltage of the stock fan by adding an additional resistor to get rid of the horrible noise.
Or do you have a better solution to lower the noise? 8)
And at the end of the day we talk about a non safety related piece of hardware with cost approx. 350 €...
Have fun
kaskade911
-
What you need to look for when replacing a fan are airflow AND pressure. If the airflow and pressure of the replacement fan are equal or higher then the replacement shouldn't make a difference when it comes to providing enough airflow. Rubber fan mounts can also make a huge difference if the noise is coupled into to casing and sometimes these can already provide enough difference to go from annoying to quiet.
-
I just changed the fan in my 1054Z. I used the Gelid Silent 5. The install was simple, and I fit a 2-pin JST connector to the new fan cable. I couldn't be happier.
-
Here's a post of mine from a few months ago on this topic with a video showing some very crude (but still better than nothing) measurements.
https://www.eevblog.com/forum/testgear/rigol-ds1000z-series-fan-replacement/msg1036879/#msg1036879 (https://www.eevblog.com/forum/testgear/rigol-ds1000z-series-fan-replacement/msg1036879/#msg1036879)
I ended up putting the Sunon back in since, as George H.W. Bush would have put it, it 'wouldn't be prudent' to just arbitrarily decide to reduce the design-intent cooling on a complex piece of electronic gear.
-
I just changed the fan in my 1054Z. I used the Gelid Silent 5. The install was simple, and I fit a 2-pin JST connector to the new fan cable. I couldn't be happier.
Just changed mine the same fan + connector. Much better.
-
My DS1054Z has just gone out of warranty so I've been thinking of doing this mod.
I just looked up the specs of the Gelid and Sunon fans. Overall the specs are very similar. The airflow is identical, the Rigol's Sunon has 25% more pressure.
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-ds1000z-fan-noise/?action=dlattach;attach=322194;image)
You could compensate the pressure drop by removing a couple of bars from the grille on the case. :popcorn:
-
For comparison: The Fractal Designs fan that some people have been using. The Specs are much worse:
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-ds1000z-fan-noise/?action=dlattach;attach=322196;image)
-
I just changed the fan in my 1054Z. I used the Gelid Silent 5. The install was simple, and I fit a 2-pin JST connector to the new fan cable. I couldn't be happier.
Just changed mine the same fan + connector. Much better.
How happy on a scale of 1 to 10? :popcorn:
-
Ok, I but the bullet and ordered a Gelid fan.
(I might look for a rotary encoder as well. Might as well look at that while I'm inside there).
-
Some people have changed the Multifunction encoder for one that has detents. I'd like to know how these detented encoders work with functions that require fine tuning, like Cursor positioning, trigger holdoff settings, the little pop-up keyboard for naming files, etc.
-
When I put a Radio Shack Sound Level Meter using fast C weighting about 2-3cm from the center of the display on a Rigol 1000 Series I get a 53dB reading. I get the same 53dB reading on a Rigol 2000 series. fwiw, on a Siglent SDG2042X at the same distance from the center of it's display I get a reading of about 50-51dB. Quieter fans would be even nicer but none of these readings are an issue, IMO.
-
Some people have changed the Multifunction encoder for one that has detents. I'd like to know how these detented encoders work with functions that require fine tuning, like Cursor positioning, trigger holdoff settings, the little pop-up keyboard for naming files, etc.
For me, the detented encoder mod works like a charm, also when entering continuously tunable parameters. The detents are pronounced enough to make the discrete selections much easier; but not too strong, so they don't get in the way when setting "analog" parameters (including those where you have to turn the selection knob multiple times). And, of course, they do not prevent you from hitting the right "fine tuning" target, since there is exactly one detent click per encoder step.
-
Some people have changed the Multifunction encoder for one that has detents. I'd like to know how these detented encoders work with functions that require fine tuning, like Cursor positioning, trigger holdoff settings, the little pop-up keyboard for naming files, etc.
There's a video here:
https://www.eevblog.com/forum/testgear/rigol-ds1054z-rotary-encoder-mod/msg737700/#msg737700 (https://www.eevblog.com/forum/testgear/rigol-ds1054z-rotary-encoder-mod/msg737700/#msg737700)
-
Also changed the fan in my DP832 - followed Guni's guide, I used a Noctua NF-A8 PWM fan and 150ohm 5w resistor. Socket on the board in both are JST XH 2.5 2 Pin Connector (Aliexpress)
I can't believe the difference it always used to be nagging me to turn it off, but now I forget it's running :)
Fan swap on DP832 and DS1054z - 10/10 would void warranty again in return for sanity :D
-
Some people have changed the Multifunction encoder for one that has detents. I'd like to know how these detented encoders work with functions that require fine tuning, like Cursor positioning, trigger holdoff settings, the little pop-up keyboard for naming files, etc.
There's a video here:
https://www.eevblog.com/forum/testgear/rigol-ds1054z-rotary-encoder-mod/msg737700/#msg737700 (https://www.eevblog.com/forum/testgear/rigol-ds1054z-rotary-encoder-mod/msg737700/#msg737700)
Well, that's better than I thought it might be, but it did look like there was some extra "granularity" to the Cursor positioning, and I'd still like to see how finely something like the Trigger Holdoff setting responds to the detented encoder.
Since I have no trouble selecting what I need without inadvertently turning the knob, I'll be sticking with the stock encoder, myself. I just don't see that the "improvement" is worth the trouble.
Talk about noise, though... pressing that button seemed really loud!
-
... but it did look like there was some extra "granularity" to the Cursor positioning, and I'd still like to see how finely something like the Trigger Holdoff setting responds to the detented encoder.
I don't see how extra granularity could result (if the right type of encoder was used), since there is one detent stop per encoder step.
-
... but it did look like there was some extra "granularity" to the Cursor positioning, and I'd still like to see how finely something like the Trigger Holdoff setting responds to the detented encoder.
I don't see how extra granularity could result (if the right type of encoder was used), since there is one detent stop per encoder step.
Ah, that makes perfect sense, thanks. I guess I'm just used to thinking of much finer resolution encoders, like 400 lines per rev. These are what, 20 steps per rev?
-
I guess I'm just used to thinking of much finer resolution encoders, like 400 lines per rev. These are what, 20 steps per rev?
12
Edit: 24
http://www.farnell.com/datasheets/1685514.pdf (http://www.farnell.com/datasheets/1685514.pdf)
The recommended on is the EC12E2424407
-
I'm just used to thinking of much finer resolution encoders, like 400 lines per rev. These are what, 20 steps per rev?
ALPS part no. EC12E2424407, 24 steps/rev.
-
FWIW, I also used the Gelid Silent 5 on mine and it seems to work fine. Huge improvement in the fan noise.
-
Is there any way of getting inside the DS1054Z without ripping off the warranty sticker or any tricks? The fan on mine is doing my head in but I don't want to kill the warranty.
It's a lot quieter than my HP 5328A was however!
-
Best way of dealing with the sticker is to use a small wedge made from PTFE and a hair dryer to soften the glue a little bit.
Works like a charm!
In case you do not have PTFE around, non-stick backing paper also does the job.
-
Thanks will do that. My other half has some PTFE baking sheets ;)
Edit: and a hair dryer although I'll have to wait until she's out before I steal that
-
Finally did something about the fan noise, it was bugging me and I was just lazy and let it go until today. I kept the stock fan but replaced the screws that held the fan in place with some rubber mounts:
(https://static1.caseking.de/media/image/thumbnail/luvi-007_luvi_007_1g_285x255.jpg)
and put a 22ohm resistor in series with the fan. I experimented with various resistances and 22 was what I was happy with - it dropped the current from 50ma to 40ma and reduced the airflow by a small amount, but the noise went down. I think the majority of the noise is not coming from the "fan", it is caused by turbulence/air flowing through the cutouts in the shield and case and there is not really a whole lot you can do about it short of chopping large holes in the shield/re-engineering the whole airflow path. What will quiet it down is greatly reduced airflow speed but of course that means less cooling. A typical fix I have used when I have the space is to use a very oversized fan and spin it at low speed, but there is just not room for that in this scope without major surgery.
-
I cut away the stock fan grill and installed one of those wire grills together with a Gelid Silent 5.
But that *might* void your warranty ;D
-
I cut away the stock fan grill and installed one of those wire grills together with a Gelid Silent 5.
But that *might* void your warranty ;D
Why would a warranty-voider even bother with a grill? :popcorn:
-
Why would a warranty-voider even bother with a grill? :popcorn:
Since I had it in stock: Why not. I don’t know if it makes any difference in shielding so maybe it is not really needed...
-
My DS1054Z has gone out of warranty so I've been thinking of doing this mod.
I looked up the specs of the Gelid and Sunon fans. Overall the specs are very similar. The airflow is identical, the Rigol's Sunon has 25% more pressure.
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-ds1000z-fan-noise/?action=dlattach;attach=322194;image)
I FINALLY got around to changing the fan and, jeez, what a difference. The DS1054Z is almost silent now. Unbelievable.
I don't know how many cents Rigol is saving on their fans but it's a few cents too many.
If anybody's scope is more than a few months old (ie. past the early death phase) then I totally recommend this. I wish I'd done it sooner.
The only difficulty was two minutes with a soldering iron splicing the Rigol plug onto the new fan (the new fan was three wire, the Rigol fan is two wire). Opening the case, etc., was simple. All together it took about 15-20 minutes to do.
PS: Yes, the part number printed on the Rigol fan is exactly the part number shown above.
-
I don't know how many cents Rigol is saving on their fans but it's a few cents too many.
Sunon fans are actually not that bad. There would have been a lot cheaper options I guess :)
-
I did an extensive evaluation of 2 very silent fan mods for DS2072A on a dutch forum (circuitsonline.net): https://www.circuitsonline.net/forum/view/137540 (https://www.circuitsonline.net/forum/view/137540)
My choice was a Noctua fan as the most silent choice with acceptable air movement. This should also work for the DS1054Z
-
I read a lot about lowering the fan noise but I can't find anything usefull about somebody who actually put (temporary) a temperature
sensor on the pcb with some long wire to measure the difference of temperature on boardlevel before and after the mod.
This is what I'm really interested in. Why isn't there an engineer or hobbyist on this forum who had the same idea and actually did it?
-
The DS2072A has a 60mm fan and the DS1054Z a 50mm fan. There are much less 50mm fans.
-
@Karel: That is one of the things I did (temp sensor on the PCB, see page 3).
@Qw3rtzuiop: My second choice was Noiseblocker, which actually has a silent 50mm fan in their product range.
-
I find it amazing that an instrument manufacturer puts in the effort to design a nice scope but ends up crapping all over it by including a very noisy fan! |O
-
I find it amazing that an instrument manufacturer puts in the effort to design a nice scope but ends up crapping all over it by including a very noisy fan! |O
I think until recently, fan noise was just not on T&E vendors radar.
Look at the various threads on this site and you'll still see a wide spread of opinion about fan noises. People used to the old-school lawnmowers welcome the new "quiet" T&E while the others (including me) who lived through the change for PCs from absurdly noisy 10+ years ago to now quietly moving 100s of W out using big, slow fans, basically expect T&E manufacturers to pay attention to the design of their active cooling solutions, with noise being a major requirement.
The good news is that many vendors have made great progress, especially in their affordable gear which is more likely to end up in low-noise office or home environments (vs. noisy labs for expensive gear). The new Keysight supplies with a large thermally controlled fan are fairly quiet. The RTB2004 has a very carefully designed airduct and a big rubber support for the fan for reduced vibration and better flow leading to very acceptable noise in everyday conditions.
However, in both cases, they still missed the mark. The Keysight fan still runs when the supply is turned off (!) and the RTB2004 fan, while thermally controlled, has very little range, so its noise floor is probably 4-8dB higher than it could be with a bit of engineering attention (i.e. a few line of code).
I am not sure about Rigol, but I will continue bugging the T&E guys about fan noise until they are appropriately quiet.
-
I am not sure about Rigol, but I will continue bugging the T&E guys about fan noise until they are appropriately quiet.
I'm with you on that - bug em until it becomes a focus.
Besides being way too expensive, I'm actually quite disappointed that the new Keysight benchtop dc power supples are not designed with a passive cooling solution. I mean, they brag about their new highly efficient power design but still add a fan in a 30 W linear power supply which could easily dissipate all that heat through a passively cooled design, even in the most confined environments.
https://www.keysight.com/en/pd-2584632-pn-E36102A/dc-power-supply-6v-5a-30w (https://www.keysight.com/en/pd-2584632-pn-E36102A/dc-power-supply-6v-5a-30w)