Vacuum Theory

Hi rwindbigler,

In your drawing, it appears that the vacuum pumps (VP) are in series. Is this true?

The time it takes to draw a vacuum to a certain level is increased by increasing system volume, which you have done. The rating that determines how quickly a system can be evacuated, a VPs capacity, expressed as a flowrate.

What are the capacities if the pumps?

What is your estimate on how much you have increased the system volume going from the 12 units to 18 units?

Best Regards,

Mike

First, I apologise for the drawing it is a rough one. The pumps are actually in Parallel. The piping is overhead and is also a 4" pipe with 3" pipes coming from the pump straight up and saddled into the 4" header (there are also check valves and hand valves for isolation of individual pumps). the two smaller pumps are rated at 750 cfm each and the larger pump 1,000cfm, We increased by about 30% volume of the original size. Also it was asked in another post what (18" 20 20") meant that should have read "18 to 20 inches of mercury" The gauges we use on our presses are in that scale. The pumps are set at 25 inches mercury.

Thanks for your interest

rwindbigler I have used tanks on packaging machinery with great success. You need to size the tanks properly and have a beginning vacuum level in excess of your requirement otherwise you will increase the time it takes to evacuate due to the larger volume of the system. Good luck. J.Conway

I do not think you are pulling a true vacuum in the correct sense. If you are pulling water and steam then a jet ejector might be a better choice.

Header sizing can be assessed by putting a vacuum gauge at both ends of it. If there is a significant difference between the readings as the pumps are chugging away, then yes, it needs to be increased in diameter.

However, the vacuum level at any moment is related to the boiling temperatures of the fluids in the autoclave chests. If the material were only water, then the vacuum level at, say 90degC would be a lot lower than the vacuum level at, say 15degC. The actual figures can be gleaned from a set of Steam Tables, an essential tool in the process engineer's toolbox.

What is wrong with disconnecting the vacuum pumps from the header and running each chest to its own vacuum pump? At least this would improve operational flexibility, thereby improving throughput, and would give an indication as to whether there is a problem with one particular autoclave chest, or with all of them. If it is one, then it could be a maintenance issue; the thing might have a leak. If it is all of them then it could be a process design issue; the equipment isn't capable of doing what is being asked of it in that configuration. In either case, it is difficult to see from here.

Why all the vacuum pumps are connected in series in the diagram is a complete mystery, and may indeed be the process bottleneck. Please try reconfiguring them in parallel, and then report back to the forum on the changes.

_{<Subscribes to thread>}

I think that in this case it matters little whether the vacuum pumps are in series or in parallel but if given a choice I would say that parallel could be beneficial as this would make a better balance and due to the fact that this is extraction rather than reducing the pressure. You must differentiate a vacuum cleaner does not need a positive pressure pump it can work with a fan - most domestic ones do. There is a difference in pulling a vacuum and holding a vacuum. Working at reduced pressure means having the capacity to cater for leaks and non-condensibles which can be small and a very small pump is generally needed for this. To remove a greater volume and keep removing it the jet ejector comes into its own here. In this case multiple pumps of varying size can be used to control the output. These could be brought into use or removed as demand varies which appears to be the case here.

Jet ejectors are also very forgiving and will accept all sorts of rubbish.

If the vacuum pumps are in series then £5GBP says that only one of them is putting in a full day's work and that the others are wastrels; as the pressure drops along the chain, there is less work for the second, the third, etc., to do.

If they are in parallel then there is scope to sweep a greater volumetric flowrate through the system, which is sort-of-where the original poster wants to be.

To pull full vacuums the series can make a difference but generally the pumps vary in the series as the amount to pull gets better down the line we are talking about vacs of better than 28" or 29" here . Hence the use of very fancy items such as the mercury vapour pump after more mechanical devices. But when using them as a extractor as is the case here parallel is probably the best choice then they will not be fighting each other.

I don't know anything about this sort of thing so I'm just posting to get educated.

What do you mean by an 18" 20 20" vacuum?

What is the duty cycle on each of the chambers? Could you add valves and simple reservoir tanks large enough to "supply" the needs of an individual chamber for a full cycle without additional pumping:-

The valve would also enable to isolate each tank in order to drain it regularly.

Am I talking about adding the cost of an empty beer keg (If you used 50 gallon oil drums it would get pretty noisy) to each station here, or am I in the wrong ball park.

The process is the removal of by-products by suction. The use of a vacuum pump to get the suction is ok - I only question the choice of the type of pump but any will probably work OK.

Your query re the Inches of vac tells me that you are not fully au-fait with vacuums but you could also work in MM. A normal 200L drum would probably go flat at the levels used here. Beer Kegs would be OK.

Vacuum levels are often given in the height-equivalent of a column of mercury. 760mm (29.92") of mercury is regarded as a standard sea-level atmosphere.

Note that barometric pressure correlates well with local weather conditions, which is why the mercury column barometer is so common.

Header is too small - either increase to 6" or place 65 HP pump at the opposite end of the header.

A "receiver" will increase the speed of the initial draw down to a level where pressure is equalized in the system but will have no impact on the time it takes to reach 20" Hg UNLESS the receiver pressure prior to draw down is less than 20" Hg. (Operating a water seal vacuum pump at less than about 25-27 " Hg will boil the water out of the seal and make really nasty noises.)

I like the Idea of placing the 65 HP pump at the other end of the header. I have always used looped systems in my other plant's. I will pass this on to my management team and see what they think.

Just do it!

we used large storage tanks. for capacity .. use larger pipe at prod. line for fast initial draw of atmosphere,,or qwik reaction ......... tank is for volume.... so pumps can recover faster because inches of vac wasn't lost like in a small volume system we ran inline formers and rotary formers for plastic sheet

Adding a receiver of sufficient size to allow the pumps to "recover" between vacuum demands as mentioned previously is the key -

If you are pulling off vacuum and steam, perhaps you want a "tall" receiver so that liquid can accumulate at the bottom, and a pump/vent at the bottom can be used to pull out the liquid from time to time - see sketch -

You may want to open a couple of valves when you charge each chest - first open the one at the chest after closing it when vacuum is established perhaps one at the extreme right (not shown in the drawing) to clear the line of any other liquid towards the receiver -

Just my two cents -

Jim Wilson

it has been veru much appreciated