Searching for a miniature vacuum pump for suction cups 1

[amrbekhit]

Dear all,

I'm looking for a miniature (has to fit into an enclosure of size 90x35x15mm or 3.54x1.38x0.59in) vacuum pump that should generate a vacuum of 200mBar. The pump will eventually power a set of suction cups that allow a vessel travelling in a municipal water pipe to cling to the pipe wall, and so would need to be capable of handling water. My aim is to have the pump connected to a solenoid valve, which is in turn connected to the suction cup. When I want to latch onto a surface, I would turn on the pump till the desired vacuum inside the cup was reached, then shut off the solenoid to maintain the vacuum, topping it up when needed.

I've done some searching and have managed to find the following pump:

http://www.burkert.com/COM/Products/MicroFluidics/Micro-pumps/7604.html

This pump can generate 200mBar overpressure and is of a suitable size. However, I am guessing that a pump that is capable of overpressurising a container is not necessarily capable of producing a vacuum of the same amount.

What are people's advice on the matter? Can I get vacuum pumps that would fit into my enclosure? Is there a better way of controlling the suction cups?

Thanks

--Amr

 

[FMJalink]

Amrbekhit,
I’m afraid you are on the start of a major engineering project. I understand from your writing that you intend to have the required pumping means on board of a vessel / container traveling in a municipal water pipeline. Looking to the mentioned micro pump, I’m afraid you will not be able to achieve the intended service.

Let me first answer your question on possible under pressure: If you download the datasheet of your sample pump, you will see on page 2 a graph showing “flow rate against suction and pressure height”. The “suction height” is the vacuum inside the cup as differential pressure to what exists in the municipal water pipeline. Please note that it is a pump and not a compressor. It is intended to pump liquid and not a gas, but as the pump is self priming, will be capable to transport (limited) amounts of gas. The title above the graph is however “Restart points”, which probably means that these values cannot be maintained. This could be queried with the manufacturer.

You will have 3 main issues to solve:
a) Forces by the water flow on the vessel / container (to be counter acted by cable or suction cups);
b) Docking to the wall (required pumping rate to move in radial direction and getting to the intended circumferential location); and
c) Sealing of the suction cups to the municipal water pipe (wall roughness, dirt, required pump cooling and leakage rate).

I will only address the latter point, as your post mentioned that item. The wall of the municipal pipeline will not be smooth. There will be some leakage through the sealing between pipe wall and suction cups. This will be the required pumping flow, probably this will be a lot more than the 5 ml/min (max) of your sample pump. In case you are able to provide a better sealing, the absence of a water flow could lead to problems as heating up and partial filling of the pump chamber.

If you have the vessel / container on a cable, why not add a hose to provide the required suction from a larger pump outside the municipal water pipeline?

 

[FMJalink]

Amrbekhit,
I have already to correct something in my reply. If the suction hose for an external pump becomes very long, your post did not mention the intended distance in to the pipeline, the pressure drop over the hose could become too high, as there is maximally 1 bar down to absolute vacuum available. The use of a venturi in your vessel / container could then be better and water from the discharge of the pump could then be used, but pumping water into a municipal water pipeline is normally due to hygienic reasons a problem.

 

[amrbekhit]

Hello FMJalink

Thanks for your reply. I just want to make sure that I have understood the points you mentioned:

I have had a look at the graph you mentioned in the datasheet. On that graph, it states that at 35Hz/24V, the pump can suck a fluid height of 600cm. Can an interpret this as the pump being capable of producing a vacuum of 600mBar relative to the outside water pressure?

You also mentioned the following:

Please note that it is a pump and not a compressor. It is intended to pump liquid and not a gas, but as the pump is self priming, will be capable to transport (limited) amounts of gas.

I do not understand the significance of this. Why is it important that the pump is capable of transporting small amounts of gas? As the system is operating underwater, I would imagine that it would be dealing entirely with liquid, unless there are air bubbles in the pipe?

The fact that the pipe wall is not smooth is an important point, and I understand that the maximum flow rate of the pump needs to exceed the leakage rate of the suction cup in order to maintain the required level of vacuum. What I do not understand is why providing a better seal would cause problems. You mentioned that this would cause heating problems. Would this be due to the pump's motor stalling as it tries to to suck fluid but can't? My original aim was to have a solenoid valve situated between the suction cup and the pump shut off once the desired vacuum level was reached, then the pump would be turned off and only turned on again once the vacuum in the cup dropped below a certain threshold. If the pump is self-priming, does it matter that the pumping chamber is partially filled?

Unfortunately, a cable or hose back to the base station would not be feasible as the aim of the project is to develop an untethered vessel. If I were to use a venturi, I would imagine that it would look something like this:

The pump would need to be continually powered in order to produce the vacuum in the suction cup unless a valve is placed just after the cup that allows the vacuum in the cup to be held.

--Amr

 

[CH5OH]

I do not know if you can find a centrifugal pump that small
but if you pump the water out of the cup, you create the required vacuum
if the leakage is to big, maibe an ejector and centrifugal pump will do the trick

 

[FMJalink]

Amrbekhit,

Graph on the datasheet
It is also my interpretation of the graph that the pump is capable to create at least initially (as the manufacturer calls this restart points) an under pressure of 600 mbar relative to the outside water pressure with a flow of 2 ml/min with a power supply of 24V on 35 Hz. Please contact the manufacturer on details, they will probably happy to help you. What I do not understand is why the discharge pressure is that small, compared to the possible suction height.

Pump and not a compressor
In the datasheet is given that it is a diaphragm pump. A diaphragm pump will have valves to separate the suction stroke from the discharge stroke. The design of these valves is different with pumps from that of compressors. With pumps the valve closing will be improved due to the liquid, which is more viscous and is heavier than gas. Also will the parts be cooled by the flowing liquid. I assume that this pump to be not very sophisticated, as it is a miniature pump, meaning the pump to be without no-load system. Also I expect these pumps to be sensitive to heat generated as they are quite small. If your sealing of the cup to the pipeline wall does not allow a leakage rate at the under pressure on the graph, will the pump at suction stroke not be able to perform a normal stroke. At the discharge stroke will the driver also not perform a normal stroke. I presume that the driver is just a spool with a magnetically affected pin mounted on the diaphragm. This would be beneficial to your application when the spool is not capable/strong enough to generate a vacuum in the pump chamber. The pulses without flow would give vibrations in the valves without cooling by sufficient liquid flow. This heating would probably deteriorate the valve seating in time. The manufacturer will most likely be able to give you details on the limitations.

Why is it important that the pump is capable of transporting small amounts of gas?
This was a misunderstanding on my side. I did not realize that the vessel / container will be fully submerged in water and cup, tubing and pump will be fully filled with liquid at the start of pumping. I thought that the vessel / container would be entered occasionally into the pipeline, were the self priming capacity of the pump could fill air filled parts of the cup and tubing on the suction side of the pump. Having an air bubble in the suction cup(s) could after docking to the pipeline wall create some problems.

Cup sealing leakage and pump capacity
I agree that the maximum possible flow rate of the pump should exceed the leakage rate of the sealing of the suction cup. If the leakage becomes very small, the valves will not be cooled sufficiently. I presume the driver to be a pin inside a spool and the pin will vibrate due to the alternating magnetic field in the spool. When the magnetic field is not strong enough to create vacuum below the diaphragm, will partially filled pump chambers not happen. If some kind of a crankshaft construction would be present, could this occur, if without a no-load system. A partially filled pump chamber would have nasty effects when continued for substantial time.

Solenoid valve
It appears possible to operate the system with a separating solenoid valve between the suction cup and the pump. I think however that the leakage rate will be larger than you think and that continuous pumping will be required, where the pump flow could be frequency controlled. Maybe a solenoid in the by-pass with water directly from the outside of the vessel / container to the pump would be more appropriate to control the system, but will request more power.

Cable, hose back to the base station and venturi
When I wrote my correction, I did not realize that you will probably will have the vessel ./container in a fully pressurized municipal water pipeline. This will allow you to use the full differential pressure (being more that 1 bar) from inside the pipeline to the outside world for propulsion and clamping to wall effects.
The idea of the venture was to use the discharge pressure of a pump, which can be much higher that several bars of pressure. The schematic would be as shown by you, but with the pump arrow in the opposite direction.
Your schematic would be as the by-pass system described above with a solenoid valve in the suction from the water pipeline.