Anyone built a silencer to the vacuum pump?

[Mikael Bohman]

Have anyone built a sound silencer to the vacuum pump?

I'm thinking of a damped ported box, maybe with 2 ports so the motor cooling still works.

(A ported box works as a second order acoustic low-pass filter)

Maybe using some standard plastic pipe, with two 3d-printed end-caps that includes the ports.

[Brynn Rogers]

I've used some compressors and vacuum pumps.    Often there are two sources of sound:

1) the vibration of the motor (linear or rotary).    To quiet this you need the good rubber isolation feet so the compressor/pump is not bolted to a sounding board.

2) air noise.   The air being sucked into the pump (on a compressor) or blown out of the pump (on a vacuum pump) should go into a foam thingy.   Bigger is probably better up to a point.

[SMdude]

When my noisy 24v pump died(rubber valve failed) I replaced it with a 240v laboratory vac pump.

It is much quieter now.

I did try an air conditioner vac pump, but it was really slow to vac out my reservoir. I need to try it again with the valve removed from the fitting I was using. These are a fairly cheap and quiet system.

[ma...@makr.zone]

Well I can tell you what doesn't really work (I planned to document this and other FAILS on my build doc):

It doesn't help much to use a dampening material as base (rubber feet and/or soft wood-fibre board). I tried several options. As soon as mounted, it will still couple the vibration into the machine frame or base.

It only helps partially with a "hammock" design:

Yes it kills the nasty bases, but the higher tone spectrum is still audible.

What helped most, was simply hiding away the pump behind the machine base frame and against the corner of the room. To do that and still be able to service the pump easily I used magnets against a steel sheet and some soft packaging "fleeze" stuck between.

It is better but not "solved".

_Mark

[Paul Jones]

If the pump doesn’t heat up much wrap a pillow around it. Also try using regular elastic bands, the smaller they are the better (as long as it still holds for a reasonable time).

What you are building is a mechanical low pass filter – passing 20 hz and below, filtering anything above. Adding mass to the pump and elasticity to the mounting is the key.

 

Paul.

 

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/260735f6-9a8b-4d19-8405-c3fc9f8a91ae%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mikael Bohman]

This is the SPL calibrated (ref 20uPa) spektrum from my pump at 10 cm, before any mods.

[Mikael Bohman]

And it is 73 dBC and 74 dBA , which gives a better idea of perceived sound level.

[Michael Anton]

I always wondered about putting the pump in a vacuum accumulator bottle.  Since the pump generates a vacuum, you can put a bottle around it, and use the pump to evacuate the air in the bottle,  this should help the conducted through air noise quite a bit.  Rubber mounts of some sort would still be required to reduce the mechanically coupled noise.  The worry with this system, is that the pump might overheat if it is using air for cooling.

[Mikael Bohman]

A first order approximation says that the sound-pressure from an point-source is proportional to the density of the medium, meaning that for each time you halving the pressure, the SPL will decrease with 6 dB. Perceptual -9 dB equals half the loudness. 

So you need to decrease the density of the air with a factor 2.8 - to achieve an perceptual halving in loudness. That is without any mechanical conducting.

[Mikael Bohman]

Work in progress. First measurement, using a mechanical filter and an acoustical filter. Red is without silencer. Both measured at 10 cm.

From 74 dBA to 48dBA , and not much energy around 3kHz  the "Human baby screaming energy" region

[Mikael Bohman]

We have a
mechanical mass-spring system == rubber bands + motor mass

acoustical mass-spring system = air compliance inside big pipe + acoustic mass inside small pipes, more known as a Helmholtz-resonator.
The resonator is heavily damped with a mixture of lamb-wool + synthetic fiber.
(If you try with 100% synthetic fiber, the effect will be isothermal compression but not much absorption.)

[ma...@makr.zone]

While I can't say that I understand what you are doing, the results speak for themselves. Impressive!

I tried to quickly read up on dBA etc. to find out what the change from 74dBA to 48dBA actually means in terms of perceived noise. Not a chance. Too many "deep experts" polluting and bloating Wikipedia there :) No simple explanation to be found, it seems.

BTW: if you were to deploy a little corner of your XMOS board to sound processing, would it be possible to create "active noise control"? Just kidding :)

_Mark

[Mikael Bohman]

Well, I actually have written a XMOS library for ANC. It is based on the Filtered X Normalized Least Mean Square algorithm.

ANC might be in the pipe. It works well with any CODEC that has a low latency mode, like (the old) Cirrus Logic chip.

To get an idea of the loudness perception you should look at Loudness curves, and translate the SPL to phon. 

But it is much more complicated with a full spectrum since the low frequency masks higher tones. If I cancel out the fundamental, the overtones will probably 

sound louder.

But, my 3D printer is going warm here, it is several parts more in my silencer before it is final. It will be a higher order system.

[TheCunningFellow]

This is how I got to 48dB BTW

https://www.aliexpress.com/item/DC24-Brushless-10W-12L-min-High-Pressure-Micro-Diaphragm-Pump-TC-50A/32490419755.html

I know it is not totally in the spirit of the topic, but it is ON topic.  $100 pump vs $14 pump = $86 delta for a 26dB delta.  That works out at $3.30 per dB (plus the fact the brushless DC pump life should be longer)

The loudest noise from this pump is the actual air path.

Lovely work Mikael, and I will try the ideas to lower my pumps noise further :)

[Mikael Bohman]

Here are some cross-sections of the full silencer. Still to be assembled and tested.

[ma...@makr.zone]

OK, now I got a taste of the Clarke's third law.

3. Any sufficiently advanced technology is indistinguishable from magic.

https://en.wikipedia.org/wiki/Clarke%27s_three_laws

One question though: what about thermal considerations?

_Mark

[ma...@makr.zone]

TheCunningFellow    
Jul 5 (9 hours ago)

This is how I got to 48dB BTW

https://www.aliexpress.com/item/DC24-Brushless-10W-12L-min-High-Pressure-Micro-Diaphragm-Pump-TC-50A/32490419755.html

Thanks for that link. Few questions.

Did you specifically search for the quietest one?

The specs say 1/8" inlet = 3.175mm. Is this the outside diameter of these:

_Mark

[TheCunningFellow]

I searched for brushless DC ones first.

Then I limited the search to ones that met the pressure flow requirements someone here (Henrik or Maltes) specified.

That then left me with only a few options all of which being quiet.  Lowest dB was not the plan - it just happened.

The ID is 3.75mm the OD is 6.3 to 6.8mm (minor and major diameter of the barbs). Nothing seems to be 3.175mm

I am fairly impressed with it compared to the cheapy ones I also bought.

I have another different BLDC one on the way from AliExpress.  I'll let you know what it is like when it arrives.

[Mark]

> The ID is 3.75mm the OD is 6.3 to 6.8mm (minor and major diameter of the barbs). Nothing seems to be 3.175mm

 

Then how do you fit the tube? My Liteplacer kit tube has ~4mm ID and ~6mm OD.

 

_m

 

 

[TheCunningFellow]

I use a short length of wider tube to go to solenoid that fills my air tank/reservoir.

If yo dont have a solenoid right near the pump then you probably best off with an inline push fit joiner to go from wider tube to the smaller.

[Mikael Bohman]

Did I mention that you need to rub everything with snake-oil to get it working ? :)

Actual, it is just a bottle with 2 necks, that happens to be folded.

The thing in front of the motor looking like a lemon squeezer is a holder for a 50 mm fan.

Den lördag 30 juni 2018 kl. 00:29:21 UTC+2 skrev Mikael Bohman:

[Mikael Bohman]

The largest problem is mechanical conducting at the moment. I changed the design a little and ended up with this spectrum.

Work very well > 1kHz

Since I use 2 plastic pipes, the mechanical self resonances of the pipes is similar. I need something that can eat energy around 500 Hz.

Taking a spectrum very close to the body surface, and a long way from the acoustic opening shows this spectrum.

Showing which parts comes from surface acceleration of the body, and not acoustical conduction from the acoustic openings.

Both plots is with the fan inside running at full speed to cool the motor.

2 low-cost solutions is to instead use sand or bitumen in-between the big pipes.

Sand has the nice property that it eats a lot of energy moving it around. I will go and by a small bag of sand I think.

To make the spectrum follow the Equal-loudness contour, I need to kill the region at 550 Hz.

Anything sine-wave below the threshold, a young human with perfect hearing cannot hear.

Den lördag 30 juni 2018 kl. 00:29:21 UTC+2 skrev Mikael Bohman:

[Mike Menci]

I came across this KNF Vacuum Pump BG31 24V DC 2480rpm 0.6A with very quiet motor
https://www.part2go.com/lab-equipment/pumps/knf-vacuum-water-pump-bg31-24v-dc-2480rpm-0-6a.html

Well you need some noise or your wife will not belive that you are working! ?

[socke]

Do you already own this pump? I can't find a datasheet for it - do you have any numbers, how quiet is it really?

[Mike Menci]

Hi, 

Yes I do - good pressure - quiet motor - but diaphragm part of pump is still laud Google Co. = KNF BG31-24vdc

See enclosed measured with iPhone abt 10 inch from pump 

[socke]

Thanks for doing the measurement! 

Unfortunately I'm more confused than before. I measured my cheap Chinese pumps (which are very loud and have too low airflow) and got lower values. It seems these phone measurements are only good for comparisons with the same app and phone...

I'm also considering the pump TheCunningFellow showed above as it is specified for 48dB. But who knows how that compares?

[TheCunningFellow]

The loudest component of the pump I have is the air inlet/outlet.  When hooked up to the reservoir the pump when on is quieter than the Meanwell SE-450-24 24V power supply if that gives you a reference point.

I don't have a professional sound level meter, so recording it next to a known noise is the best we could do to compare.

[socke]

Thanks, that seems indeed really quiet. Btw., what other pump from aliexpress do you have on the way?

[TheCunningFellow]

I just went looking for it now and can't find it.

It claimed 55dB but had slightly worse vacuum performance than the one I first got.

It was $60 USD vs $100 USD though so I thought I would try it for my manual hand operated PNP helper.

These are the 4 stores I was looking at for other pumps.  Seems $50 and up they start to get quieter.

https://www.aliexpress.com/store/1522253

https://www.aliexpress.com/store/508843

https://kamoerfluid.aliexpress.com/store/500411 / https://kamoer.aliexpress.com/store/935395

https://www.aliexpress.com/store/1332185

[Peter Chaisty]

Hi all

I use kind pumps of various sizes in some of my equipment.
They are fantastic quality and I buy direct from knf.
I have some of the smaller ones so I can look on Monday how quiet they are.
However I use some very large by comparison pumps and I use a silencer from rs components.
It makes a massive difference.
I'll lookup the part numbers on Monday.

Peter

[Trampas Stern]

My PnP uses an air pressure to vacuum converter, only is on when vacuum is needed and is very quiet.   I would much rather maintain and quiet down my shop air by putting compressor outside, than worry about vacuum pump maintenance, reliability, and sound damping.  

I use shop air for so many things, like blowing off stencils before use. As such I consider compressed air a requirement. Also if you use the Yamaha feeders then shop air is a requirement.  I even use shop air for cleaning off dust and stuff, it makes life so much easier I can not see how people work without it.  

Trampas

[Mikael Bohman]

For anything that should correlate to human loudness, it is better to use dBA instead of flat dB, at least at lower SPLs.

Two measurement of 85 dB can have a totally different loudness. Maybe you can change it in the app ?

[Marek T.]

Agree with @Trampas. Additionally compressed air allows to get blowing into nozzles (in my system I have blowing or sucking, there is not state of "nothing") what improves parts releasing while placement.

[Melitonas]

Hey,

I ordered the same BL pump from Ali. What reservoir do you use for it? 

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/818e105a-4b32-42ea-b1ff-944284eb8b63%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mark]

> it makes life so much easier I can not see how people work without it

 

Good point. It turns out that way some times.

 

Just as an example: Before I had the heat gun I was not aware of how often and for what diverse purposes I would use it. Like for gluing stuff with hot glue. One can practically “reflow” and “rework” stuff almost like with solder. Life saving for first-of-a-kind contraptions.

 

_Mark

 

 

 

                                                                                                                                                                            

[Malte R.]

 I use a different pump which also was a little noisy.

Eventually I made an isolated, wooden box for it.

I am reasonably happy with that solution; noise is no longer an issue. I don't have the equipment to measure exact db(A) though.

Find description and pictures here:

https://www.liteplacer.com/phpBB/viewtopic.php?f=4&t=145&hilit=+pump+noise#p1172

[TheCunningFellow]

Yes - venturi vacuum is great, but in some cases a large compressor is not practical.

I also have pressure for doing "puff" from that pump.

I made my set up mostly copying ideas from Malte as I can't run my 3HP shop compressor after 7pm.

[Peter Chaisty]

These are the kind of silencers I use

Then also anti vibration mounts.

I use the type with a male and female thread but they come in all kinds of styles

This makes a huge differences on my larger pump systems.

Peter

--
You received this message because you are subscribed to a topic in the Google Groups "OpenPnP" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openpnp/RxfOlMciHKc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openpnp+u...@googlegroups.com.

To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/0181947d-4c9c-41e6-a2e9-05cf05c2ae93%40googlegroups.com.

[Mike Menci]

Peter
Where do you mount the silencer to ?
Mike

[Peter Chaisty]

Hi

It goes on the exhaust port.

On my larger pumps I can screw then straight in but on smaller ones you can use tubing.

They are good in that they do not restrict airflow.

That cures the noise caused by the movement of air then anti vibration mounts cure the motor vibration issue.

Peter

--
You received this message because you are subscribed to a topic in the Google Groups "OpenPnP" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openpnp/RxfOlMciHKc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/1b775e15-7278-4a6f-a347-6c53e56781d7%40googlegroups.com.

[Trampas Stern]

Yes the first time the nozzle drops a part and goes down on some solder paste having the blowing will save you the cost of a nozzle.  I personally plugged up a nozzle because I did not have vacuum sense to know if part was picked and I did not have blowing. Even compressed air would not clean nozzle after solder paste dried in it. 

[socke]

I'm still comparing pumps and their datasheets. If I'm right, this pump has an airflow of only about 3l/min (I found only the datasheet of the newest model), compared to the pump from aliexpress which is specified for 12l/min.

This is some difference. My question is, what is a suitable lower limit for the airflow? Can we have way too much airflow? Or would it be just wasted power?

I'm planning to have a dual nozzle system with vacuum sensor for each nozzle - so it would be nice in case of a failed pick if the part from the other nozzle wouldn't fall down immediately.

Thanks for all comments so far!

[socke]

Here I mean the above mentioned KNF pump (BG31).

[Marek T.]

My nozzles are non stop blowing instead of moment when I switch the venturi valves into vacuum mode. Blowing is usable because the part is not only dropping down when the vacuum is missing (like in solution with switched vacuum pumps) but additionally is repelled by the blow.

It's good while placing but problemable while picking. Blowing air can blow out 0402/0201 from the tape sockets. Turning on the vacuum before nozzle lowering results with pcking the part before nozzle is at final down position.

Like always, some pros and cons of any solution :-)

[Mikael Bohman]

Next version:

Colling fan

Pump with rubber bands and mount.

Goes into black pipe.

2 standard plastic pipes. Sand in between. Hot melt glue to seal.

The objective is a 10$ silencer, if you have a small 3D printer.

Sound Level emitted from mechanical enclosure, taken at a distance of 1 mm from the pipe. (The SPL decreases with the distance until you reach the reverberation radius)
Acoustic opening sealed to only measure sound produced by mechanical surface acceleration.

[Mikael Bohman]

Measurement of above at 1 meter. Bottom port seal with the carpet.

Spectrum and dB(A) at 1m non-anechoic measurement.

Good enough to continue with :)

[Jason von Nieda]

This looks pretty great! Cheap and easy, which is even better :)

Jason

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.

To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.

To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/bcf00caf-a04f-4b08-9496-1f7a2149b480%40googlegroups.com.

[TheCunningFellow]

Must have been drunk when I ordered this one.

https://www.aliexpress.com/item/Mini-diaphragm-vacuum-pump-mini-multi-air-compressor/32275418738.html

Not only is it a brushed motor rather than brushless.  It is also 12V rather than 24V.  Still it is almost 1/10th the price of the other one and is almost as quiet.

Anyways - the pump itself seems to produce as much vacuum pressure by the size parts it can pick up.  Has much lower flow rate though.

I'll try to get time over the weekend to compare its absolute pressure and flow rate compared to the other by measuring how fast it empties my 1L vacuum tank.

[TheCunningFellow]

Relative performance of the two pumps.

The photo shows the output of the MPX5100DP.  Full scale 5V is meant to be 100kpa, but this is obviously not calibrated.

Anyways - the large pump is meant to do 70kpa and the small one is meant to do 50kpa.

The top line made it to 67kpa and got close to that in 30 seconds.

The smaller pump made it to 57kpa and took 90+ seconds.

The tank has approximately 1 liter capacity.

Both pumps sound about as loud when starting up, the small pump gets much louder as it reaches max pressure.

The small pump even when it got louder at the end was nowhere near as loud as the cheapy sparkfun/ebay pump unloaded.  The large pump remained quiet.

The flow rate of the small pump is enough that it has no problem picking up parts on a 503 nozzle with a second 503 nozzle attached and just leaking air.

I would say the small pump (which looks the same as one robotdigg sell) would be adequate as a quieter replacement for the sparkfun pump.

[socke]

Thank you for the comprehensive information, I really appreciate it!

[TheCunningFellow]

I just got a cheap sound level meter.

I am sure it is not super accurate - but it is some form of measurement.

The large brushless pump when connected to the air tanks is 45dBA at 1 meter distance.

The small brushed motor pump is 49dBA.

Both of those numbers jump by about 6 points if they are not decoupled from the desk well and the vibration gets amplified.

If you don't muffle the ports and point them at the sound level meter you get an extra 10+ dB

The meanwell brand 24v PSU fan is 52dBA.  So that confirms my ear based guess that the pump was quieter than the PSU.

[socke]

Last week I've got a used KNF pump from ebay with a brushless motor. The really nice thing about it is, that it's speed can be controlled by an voltage which relates to a air volume between about 5 and 25 l/min. It's a real monster at full speed, but I think I will run it at very low speed and low noise. ;-) 

Especially at low speed, the only noise comes from the pump - the brushless motor is almost not noticeable.

Maybe I will use a little Arduino to build a simple pressure control for it.

[Mike Menci]

Noise protection - 12 V DC - vacuum pump with tank (auto on/off) 

https://www.ebay.com/itm/Mercedes-Benz-E-Class-W210-W208-CLK-Central-Locking-Vacuum-Pump-2108001948/192852897988

- learn here; https://www.youtube.com/watch?v=ShvDDiXncSE&t=517s

Mike

[John deGlavina]

Currently printing this one with a 1/4" barb instead of 1/2":https://www.thingiverse.com/thing:2774371
It probably wont do that much, but worth a try.

[Allan Madsen]

Why not just use a pump from a freezer or a refrigerator?

[Marek T.]

Or silent stomatological? :-)