The Combine Forum
This is the final post I wrote on the "Push Button Combine" thread.
These posts plus some shorter ones in the thread took a lot of time and research to try and make sure I got my information and facts straight as I was writing for a foreign audience who had been led to believe that they were the original inventors and developers of all the significant advances in agriculture machinery over the last 200 years.
I simply ran out of enthusiasm and so decided to stop while [ I hoped! ] I still had considerable interest in what I was posting.
This final post was getting into the technical side of combine design which I along with my very innovative brother who I was in partnership at the time, have dabbled rather heavily in whilst developing our unique medic seed harvester.
My thanks to all those who have had the interest, the patience and stamina to read this far. The Push Button Combine
Part 17 & Final
17 / 4 / 2010
Most grain growers have a fairly good knowledge of the principles behind the various processes that are carried out during the collecting, threshing and separating processes in a combine.
However I suspect that a lot of owners / operators more or less throw their hands up in the air if you ask them just what are the design principles behind the designs of the sieve fans in a combine.
Those hands might only go up a bit when you talk about the age old garden variety paddle fan but when the discussion gets onto the principles of operation of a squirrel cage fan and / or a cross flow / vortex flow fan then a certain silence might descend on the discussion.
A few days ago I wandered around the dealer's yards that represent five makes of combines in this town.
One dealer, Caterpillar, did not have any combines displayed so I couldn't check the Lexion
I was just a little surprised at the variety of sieve fans I found on both the new models and the older traded models.
Now what I saw may not be the absolute latest and no doubt I will be corrected very rapidly if I get something wrong.
My excuse is that I no longer even try and keep up with the telephone numbered series of model identification that some manufacturers seem to delight in these days and if I did I probably would get a 9670 mixed up with a 9760 or a 7690 and etc if such things exist.
These numbers might mean something to the aficionados of that manufacturer's products but when you get at least three manufacturers all designating their models with similar and interchangeable telephone numbers it does show a truly remarkable lack of imagination by their marketing sections.
I don't know about Americans but if the car manufacturers tried to sell their entire range of models here in Australia starting with numbered models like model 62, something close to a push bike, 625, about 5/8ths of a reasonably performing vechile, 6.25.8 for the family job and then the 625-7.85 for the performance machine I doubt that the sales figures would be exactly spectacular.
Call them Cobra, Falcon, Corvette, Mustang and etc and sales would go right up and everybody would know which make and model you are talking about!
Although come to think of it, a [ hypothetical ] "Mao" model from "Great Wall Autos" does seem to lose a certain something in the translation.
I have along side of me a Buffalo Forge Fan Engineering book of biblical type dimensions with some 860 pages dealing with all aspects of air and gas movement.
Fans and fan design alone, take up some 200 pages of this book.
It was always close at hand when I spent a number of years along with my brother, designing and building an one off, ultimately very successful suction type medic pasture seeds harvester which I commented on in the Two Rotors thread.
That darn book cost me an eye watering $180 AUD in 1983 and that was at wholesale rates as it was not the very latest edition.
Now fans look bloody simple and anybody can build one!
That perhaps applies to the simple old basic but inefficient paddle fan in it's sheet steel casing.
For the rest of the fan designs, well I have news for you!
Fans both in design and in construction are a very sophisticated bit of engineering and involve some very complex formula's and calculations if you are to get an efficient and good performance fan.
There are some specialist fan design companies who do nothing but design fans and test those designs.
Then there are specialist fan builders who often have their own design teams but also build fans to designs from the specialist fan design groups and who do no other engineering except build fans and fan systems.
Very few people are also aware of the basic fan laws with regards to fan output and the power that fans require.
It is always a very good idea to know and remember the following simple facts about fans.
Provided the fan is running within it's RPM design range and importantly, has an unobstructed inlet and outlet flow and all other factors, most of which are quite important in practical use, are unchanged;
1 / Air volume changes in direct relationship to the changes in fan RPM.
2 / Pressure changes are the square of the fan RPM changes.
3 / HP required is the cube of the fan RPM changes
If we use a practical combine based example where the combine fan RPM is increased by say 20% or 1.2 times which is well within the fan RPM range used by combines.
1 / Air volume will increase in line with the 20% fan RPM increase = x 1.2
A 20% increase in air volume for a 20% increase in fan RPM's
2 / Pressure would increase as the square of the RPM change; = 1.2 x 1.2 = 1.44
Pressure will increase by 44% for a 20% increase in fan RPM's
3 / HP required will increase as the cube of the RPM increase;
= 1.2 x 1.2 x 1.2 = 1.728.
HP requirements will increase by 73% for a 20% increase in fan RPM's
The converse is of course also the case with fan RPM reductions.
That sort of HP increase / decrease has all sorts of connotations for the drives on combine fans but other factors also come into the practical side like the amount of resistance to the fan airflow that exists due to the sieves and of course, just how much resistance to the airflow the mass of threshed material on the sieves has.
There are also some variations on those basic Fan Laws due to the different design approaches used in fan design but the basic underlying thrust of the fan laws holds for any particular fan design.
There are three basic types of fans.
First is axial flow type of fan where the airflow is along the axis of and through the fan.
The common multi bladed domestic cooling fan and the vehicle radiator fan are examples but the axial flow type of fan also includes some of the most sophisticated fans around, the multi-bladed gas turbines used in aircraft and in the steam and gas powered electricity generating power stations.
Another example that is familiar to rural Australian's is farm reservoir water and / or well / bore pumping windmill. The Windmill Journal
[ Being the next to Antarctica as the driest continent on Earth, the ready availability of water in Australia has a significance that for North Americans and Europeans with their immense and readily available supplies of water, find hard to grasp. ]
With the more simple versions of this axial flow fan, pressure rises and the controlling of the rise and flow evenness with the twisting or torquing of the airflow from the rotation of the axial flow fan makes for some rather difficult engineering to get evenness of air flow across the entire sieves for such a fan design in combines.
This rotation and unevenness of the airflow can be corrected by using deflectors and guide vanes to straighten the flow of air but this then increases drag on the airflow and significantly reduces the overall efficiency of this type of fan design.
The type of Axial flow design fans that we use in everyday use can be choked right off without any ill effects to the fan.
But that characteristic also poses problems when the sieves become overloaded and the air flow then just shuts down and blocked sieves may be the result.
However, despite the inherent design problems in creating a smooth airflow from this type of axial flow fan under sieves, I think Claas at least, still used a bank of axial flow fans under the sieves in some of their combine models.
For interest, just to give a few examples of axial flow fans and the sizes they can and are manufactured to; Wind Tunnel Fans
The next is the very common and popular radial flow or centrifugal flow fan type where air is drawn into and flows into the centre of the fan impeller and is then through centrifugal force, thrown out through the fan blades, ie; flows out of the fan in a radial flow.
The air is contained and directed by the fan casing which usually takes the form of an increasing in radius scroll as the casing contains and directs the air flow around the expanding radius of the scroll into an outlet.
This basic fan design can range from the most simple 4, 6 or 8 bladed paddle fans that were and still are found in nearly all of the older combines and are still found in some of the latest combines right up through to the huge in both diameter and capacity, centrifugal mine ventilation fans or power station boiler and exhaust extraction fans that take thousands of HP to drive.
The centrifugal fan design type encompasses a huge range of fan designs using the basic centrifugal or radial outflow design.
It includes the narrow cased high volume and / or high pressure fan types that are not that dissimilar to the centrifugal water pump in design and operation as well as the wide cased, very high volume fans for aeration and ventilation.
The difference between the centrifugal water pump and the centrifugal fan is just that the centrifugal water pump is designed to operate with an incompressible fluid whereas a centrifugal design fan, as do all fans, operate with a highly compressible fluid, ie; a gas or air in our case.
The paddle fan is cheap and easy to build, reasonably efficient, gives a good flow of air that with a little tweaking of the casing / scroll design can give an even airflow distribution right across the fan outlet and therefore is very good for the evenness of air flow under a combine's sieves.
Which is why for some 150 years or more it has been the most used design of fans for an innumerable number of applications of every conceivable type that require the movement of air.
The useable width of the paddle bladed fan is limited by the amount of air that can flow in through the open end holes of the fan scroll casing before that air turns into a radial flow out through the fan blades.
Too long a fan of this type and sufficient air will not reach the centre of the fan leaving a dead spot on the sieves and causing big disruptions to the airflow from the fan under the sieves as the air tries to flow across to the low pressure region to even out the pressures.
To overcome this, some combine manufacturers split their paddle fans into two units on a common shaft with a large opening in the middle of the fan scroll housing to allow air to flow into the centre of the fan and provide an even air flow right across the sieves.
The simple paddle bladed fan also is not particularly fussy about the accuracy of the scroll of the casing so considerable dimensional intolerances can be accepted when being manufactured before a noticeable falling off in performance can be detected.
As with all fans, the clearances of the cut offs to the fan wheel where the fan air flow changes from the outlet to the start of curve of the scroll can make a big difference to fan performance but once the air starts it's journey around the scroll towards the outlet, paddle bladed fans are not very fussy about clearances.
The other check to be made is to have the minimum of clearance around the end of the fan blades and between the blade ends and the end housing of the casing / scroll.
Leakage around the blade ends or uneven spacing between the ends of the blades of the fan and the scroll end housing will allow considerable air flow around the blade ends and allow variations in air flow under the sieves.
Centrifugal fans can and do also have neutral, forward curved and backward curved blade designs, all of which give different pressure rises and volume flow characteristics so that a fan design can be selected to achieve certain air/gas flow characteristics.
JD some time ago in some of their models used a bank of closely spaced and very sophisticated narrow cased centrifugal fans mounted on a common shaft under their sieves.
NH still appears to be using the old bog standard simple paddle fan under their sieves if the what I assume is a late model NH combine I looked at a few days ago represents NH's current fan installations.
A version of the centrifugal fan is known as the Squirrel Cage fan named after the wheels that I assume tame squirrels [ no squirrels in Australia except in zoos. ] and tame rats, mice and etc get inside to run and exercise.
The Squirrel Cage fan has a large large number of closely spaced, long, narrow, curved forward blades set well out on the periphery of the fan wheel.
This particular version of a centrifugal fan design if properly designed has some excellent high volume flow characteristics and some good pressure rises as the the fan RPM's are increased to handle heavier sieve loads.
JD appear to be using a coarse squirrel cage type of centrifugal fan at present in [ some of ? ] their latest model combines.
Squirrel Cage fans can also be found in small versions in the combine and tractor Aircon cabin fans, in some domestic water cooled aircon units and in a number of other very high volume, low pressure applications particularly in the structural cooling and heating industries.
There also some very sophisticated mixed flow fans where the axial flow design is mated with the centrifugal design to give a very high efficiency fan impeller with big pressure capabilities and good volume flow rates.
Turbo chargers both in the blower impeller and the exhaust turbine use mixed flow impeller designs and with rotational speeds up to over a 100,000 RPM in some instances, some very hefty pressure rises can be achieved from turbo chargers.
Allis Chalmers techs once checked our now long gone 7045 tractor as we were having fuel pump problems which subsequently were discovered to be caused by excess paraffin wax in the distillate.
On cold mornings [ like cold for us is 5C below freezing for a couple of hours! ] the paraffin wax in the diesel fuel was freezing solid and the thin sheet of fuel which acted as a lubricant for the closely machined innards of the pump was scraped off from between the rotor and housing allowing bare metal to bare metal contact and the resultant scoring of the pump rotor and housing and a buggered pump.
All of which led to some loud language and some table thumping in an interesting meeting with the technical marketing managers, the guys between the refinery and the wholesalers, of the 27 oil companies in Australia at that time.
I had a lot of info from a lot of research over the previous few months on what turned out to be a widespread and serious problem for farmers, truckies, earth movers and etc so I very unexpectedly and very nervously got the job from our farming organisation of taking some of those tough oil company technical marketing managers on.
Subsequently and very quickly the excess paraffin wax problem in diesel fuel disappeared as the oil companies did not like being told face to face that they were responsible for a lot of severe damage to diesel engines due to their fuel policies and that they were ripping us off plus some TV exposure.
Got pretty rugged at times in that meeting but strangely I had quite a few of those oil company managers who supported me and the two other farm organisation guys who came along with me for support.
Back to the AC test.
Can't remember the exact numbers but the turbo charger on the 7045 was cranking up to something in excess of 30,000 RPM [ I think. It could have been more! ] at full bellow on the PTO dynometer they brought along and the pressure rise from the turbo idle to full power was an astounding 28 PSI which made the AC tech's eyes bulge.
No turbo waste gate on that 7045 either.
Centrifugal fans including squirrel cage fans all suffer from one major drawback although one that a good combine operator will rarely experience.
That is they can be choked off when they reach a critical pressure depending on their design.
They just spin away and totally quit performing with no pressure as well as no flow.
Just the same as you can turn off or throttle a centrifugal water pump with a tap on the outlet or inlet.
The same factors apply to a centrifugal fan.
So if you achieve some heavily loaded or over loaded sieves then the fan if it is of the centrifugal or axial design will just quit and do it suddenly as was deliberately demonstrated to me by a friend in his big White combine some years ago.
His fan pressure monitor just said of the fan pressure "I quit" as he drove hard and deliberately overloaded the sieves, something that a decent sieve fan design should never allow to happen.
The third basic design of fans is the Cross Flow fan or sometimes known as the Vortex Flow fan or Tangential flow fan.
This particular fan design makes hardened engineers throw up their hands as they try to get a handle on the way in which the cross flow fans work.
Cross flow fans are increasingly found in combine sieve systems as they have a number of very desirable characteristics from a combine engineer's view point as well as from a combine operator's point of view.
AC with the advent of the L series were possibly the first combine manufacturers to use the cross flow fans followed by MF in the late 1970's with a somewhat crude version but nevertheless a cross flow fan in the last of the MF Australian built combines, the previously posted "11 second combine".
The cross flow fan was first patented by Mortier in 1893 so the basic design is quite old.
The cross flow fans look very much like the Squirrel Cage fans in their design with a large number of long, narrow, curved forward and closely spaced blades arranged around the periphery of the fan wheel.
However the way the Cross Flow fan works has no relationship with the squirrel cage fan or any other fan design.
The air flow in a cross flow fan is through the rotating blades right across the width of the fan and into the centre of the fan.
There is no central along the axis, infeed of air as it applies to the centrifugal and axial flow fans.
All the air flow enters the fan through the rotating blades along it's length.
Near the outlet duct of the fan a very concentrated, intense and very high speed rotating vortex is established which is partially within the fan wheel blades and partially external to the fan wheel blades.
This steering vortex runs the length of the fan outlet duct and the blades of the fan traverse through the vortex as they rapidly rotate.
They do this without disrupting the vortex structure.
This steering vortex then directs the airflow into the exit duct of the fan and a strong outflow from the fan to under the sieves is established.
For some good illustrations and information on the air flow in a cross flow fan, the paper; NUMERICAL AND EXPERIMENTAL INVESTIGATIONS OF CROSS-FLOW FANS
is about as good as I can find.
There are quite a few research papers on cross flow fans as there are still a lot of not fully researched characteristics to be sorted out in Cross Flow fan design but unfortunately most of these cross flow fan research papers are behind pay walls.
From this you may have gathered that there is till a considerable amount of art still involved in cross flow fan design rather than the use of available hard concrete data.
Cross flow fans have some very interesting characteristics which makes them very popular in computer cooling fans and other areas where small diameter fans with good volume output are required.
As the cross flow fan takes air in through it's periphery right along it's length then there is no theoretical limit on just how long you can make a cross flow fan.
The air just keeps right on flowing in.
As the air inflow and therefore the air output is even along the whole length of the fan, it is ideally suited to sieves and as the various models of combines are changed in their respective sieve widths it is only necessary to shorten or lengthen the fan and it's scroll housing to fit the changed width of the sieves.
Manufacturing is also very easy as the one size, except for width fits all with no air flow problems arising from increasing width.
There is another very desirable characteristic but with a potential and literal twist to it.
Cross flow fans unlike all other fan designs, will continue to increase pressure if their outlet is blocked with say blocked sieves until such time as they overload their drives or in the worst case have a structural failure and wind themselves up into a knot.
That is why when you look at the top of the outlet fan duct panel in the R series Gleaners, there is a full width, approximately 50 mm wide slot across the entire top panel of the fan duct and a big sign saying "keep slot clear".
If you get blocked sieves with a cross flow fan and the air flow from the fan has no escape opening to relieve pressure, the fan will slip the drive belt or destroy the drive or wind itself up into a long tangled mess of metal.
That slot is a pressure relief slot for just such a blocked sieve circumstance.
The other advantage of the cross flow fan is the small overall diameter of the whole fan system for a lot of volume and pressure.
Another simplified problem for a combine designer.
He doesn't have a bloody great big fan to fit into too small a space.
The down side of cross flow fans from the combine designers viewpoint is the designing of a cross flow fan that will actually perform with the characteristics that are required.
The design of the impeller is very important as it is with all types of fans.
However with cross flow fans the design of the scroll housing is of the utmost importance as the velocity, location, size of and position of the critical steering vortex hinges very much on the design of the scroll housing in both the inlet and the outlet ducts areas.
Also the accuracy of the build is very important to get even airflow right across the sieves.
I suspect that some combine manufacturers have not quite got on top of this just yet.
I had some recent experience while messing around with a cross flow fan from one of those MF combines of the 1970's.
A very good personal friend, regarded by some of the world's top class international competition glider pilots as arguably the best FRP [ Fibre Reinforced Plastics ] glider / sailplane repairer and refinisher in the gliding world, wanted a dust extractor so for starters, we had a look at a cross flow fan from an old wrecked MF combine.
That fan was crude and needed about double the number of blades in it's wheel to be any good but it did the job in those late 1970's combines.
When making up the scroll and ducting, I copied as much as possible the original scroll of the fan, all of which I mounted on the steel bench top.
Over many days of messing around, I then found that even only a few millimetres adjustment in the fan duct and in the scroll configuration in a number of places created large changes in volume output.
It was a good lesson in the difficulty of designing a cross flow fan and no, I eventually figured it was going to be a long and costly business to start a new design dust exhaust system from scratch so my friend just brought a well proven make of dust extractor that worked from the get go.
Looking at cross flow fans in combines like the Gleaner series, I am still trying to figure out if the Gleaner designers of the original cross flow fan used in the L series were geniuses to design that fan which is still unchanged in all important dimensions in the latest Gleaners, nearly 30 years on.
Or were those designers just plain dumb lucky to crack the key to that cross flow design that has proven to be so good that it has been nearly unchanged and lasted all those years right across the various Gleaner designs since.
And they never had computers of any sort to do that designing on either!
Another intriguing feature of the Gleaner fans is the use of a throttling board to control the air flow into the constant speed fan, a huge benefit in simplicity and engineering.
That throttle board has to also be set to some close tolerances and clearances to avoid uneven airflow across the sieves.
What also intrigues me is the fact that no other manufacturer uses such a throttling board system when they have installed a cross flow sieve fan.
It's extremely simple, works and works well so why isn't it used elsewhere on cross flow fan systems for airflow volume control?
There is another little story on various cockups that occurred with that 11 second combine.
When MF finally got the prototype and preproduction runs of that combine sorted, they put it into production.
As soon as the production combine hit the fields, there were screams from the customers that the [ cross flow ] fan and sieves were, to put a not too fine a point on it, shxt!
As the fan and sieves had performed extremely well during final testing of the prototype and preproduction combines there was considerable consternation in the executive and designer ranks.
A careful check of every aspect, jigs, build, assembly was carried out and everything was right up to the specs as determined in the prototype and preproduction models.
In some despair and with hands being thrown in the air in frustration, a couple of executives decided to wander down to the paint shop.
There they discovered a painter who with much muttering about the stupidity of those further up the assembly line had taken to standing on the fully assembled fan and fan scroll housing so that he could paint a relatively inaccessible part of the combine way up in it's guts.
That of course, distorted the scroll by a few millimetres and consequently the fan failed to perform and the grain cockies got lousy samples from their brand new combines.
Next up maybe will something on sieve systems but a lot shorter than this effort.