Sciencemadness Discussion Board

Drum dryer design

denatured - 23-8-2010 at 06:37

Anyone knows where I can find designs for drum dryers (and similar equipment)? I found diagrams for it in the "Handbook of industrial drying", but they are just diagrams. like this one:

Will I be able to find more detailed information on the construction?


Magpie - 23-8-2010 at 07:41

Are you looking for industrial size equipment or laboratory (bench) size? Ie, what diameter and length of drum do you need?

Or are you just looking for information on industrial equipment?

gsd - 23-8-2010 at 07:53

This diagram gives the very basic arrangement of a drum dryer.

The hollow metal drum is supported horizontally on hollow shaft.
The Material of Construction of the drum, its surface finish and its mechanical design as a pressure vessel to withstand the rated steam pressure from inside without any bulge whatsoever are extremely critical design parameters.

Steam is fed into the drum through the hollow shaft from one end via a rotary joint (either a mechanical seal or gland packing)

The condensate is removed by a "L" shape dip pipe inserted from other end of the hollow shaft also thru' a rotary joint.

The drum is rotated using a gear and chain-sprocket arrangement. Some sort of mechanism such as stepped pulleys / taper pulleys or frequency drive etc is used so that the drum speed can be adjusted if required. BTW the drum rotates very slowly. I have not come across a dryer with more than 5 rpm speed. About 1 to 2 rpm is a norm but that totally depends on your drying process requirement.

A feed tray is provided at the bottom. The slurry / liquid level in the tray is so adjusted that it just touches the bottom of the drum.

As the drum rotates, the slurry is picked up by the drum surface and due to steam heating, water evaporates leaving dried solid on the drum surface.

This solid is scrapped off using so called "Doctor knife".
The selection of material of construction of this knife and also the design of support of this blade is quite critical as improper selection of MOC will damage the drum and if the blade is not properly supported and the equal pressure is not applied along the entire length of the blade then the dried material will not be scrapped off properly and it will contact the liquid.

The scapped off solids are carried away by using a chute or a screw or a belt conveyor etc.

Generally a hood with ID fan is provided on the drum to take away water vapours.

Several accessories are used such as heating arrangement for the feed tray, cooling for the product etc.

Given this basic arrangement, several variations are possible such as double drums, feeding using applicator drum, top feed, spray feed etc etc.

This is the hardware part.
Now given the solid content of the slurry and required moisture content of the product and the product capacity, the software part consists of:

1) Fixing the Drum surface area (diameter and length)
2) Speed of rotation
3) Steam pressure and flow rate

If you are designing a continuous process dryer then this is not as simple as it sounds as industrial drying is an extremely tricky unit operation. Except for few hand full products, a design of (any) dryer is not undertaken without taking actual trials on the pilot scale models. As the drying characteristics as will as heat transfer coefficient for the given system can not (as yet) be theoretically predicted.

However a batch type of dryer where product capacity is not critical can be made to work if enough flexibility is built into the design.

Any standard Chemical Engineering textbook will give you the relevant details. And of course googling helps.


denatured - 23-8-2010 at 08:18

Thanks gsd for your informative reply.

Magpie, I'm searching for information so I can make a small drum dryer.

I'm thinking of making something small like 50 (48) cm diameter and 30 cm long and use oil (vegetable) as heating medium and a heating coil (1200w). I don't know if anyone ever used oil instead of steam in drum dryers, I think steam is just convenient in a large scale setup and that is why they use it, what do you think?

Thickness of drum wall: why do they use 2-4 centimeters (to store heat? to withstand steam pressure?)? what happens if I use 2 millimeters?

It won't be used continuously (batches), so no need for screw or a belt conveyor.

MOC: I think 304 steel will work for the drum but now I'm not sure about the blade, should it be made from a different steel?

gsd - 23-8-2010 at 08:52

1) Oil can be and is used as a heating medium. However in that case you will need to distribute it evenly over entire inner surface of the drum. This can be done by inserting oil feed pipe along the length of the drum with very small spray holes all along the length. The point is the top surface of the drum must be adequately wetted as that is where the maximum drying takes place. Also the oil must be sucked out from the bottom of the drum using pump.

The steam is convenient because both these problem with oil - distribution and removal are not applicable here due to it being a pressurized vapour.

2) Very high thickness of drum wall is required because when you pressurize it from inside by steam, it should not have even a slightest bulge as it will hinder proper scrapping of product.

In your case of circulation of hot oil the drum is not pressurized so such a great thickness will not be required. However you will still need a perfectly circular drum with ground finish which means you will have to machine it on a lathe and then finish the surface with grinder. So the initial thickness will not be less than at least 15 mm if you have to get about 5 mm final thickness.

3) MOC totally depends on your product and slurry. Only thing I can add here is the MOC of blade MUST be softer than MOC of drum. Remember blades are expected to get worn-off the drum is not.

Generally depending on the compatibility, brass, copper, aluminium etc are used for blade material.

Industrially harder materials are used for blades but in that case the drums are plated with hard chrome / stellite etc.


bbartlog - 23-8-2010 at 08:59

Steam is used in an industrial setup because of its ability to transmit large amounts of heat quickly (by condensing, freeing up the heat of vaporization). You could use hot oil, along with some way to circulate it, but you would not achieve nearly the same heat transfer (and thus drying throughput) as with pressurized steam. Obviously, throughput may not be a big issue for you - just spin the drum more slowly. Working with pressurized steam is hazardous and brings in special design considerations.
Likely the thick drum wall is to resist bowing of the drum wall under pressure. For the dimensions you are talking about, 2mm thickness would likely be enough for unpressurized oil.
Not sure about the doctor blade. Seems dependent on the characteristics of the material being dried, i.e. is it abrasive and how easily does it separate from the stainless drum? If the blade has to literally scrape the drum then maybe a different material will be required, but if it can function at a slight standoff then I'd just use stainless.

Magpie - 23-8-2010 at 15:29

Quote: Originally posted by denatured  

It won't be used continuously (batches), so no need for screw or a belt conveyor.

I want to compliment gsd and bbartlog for their most informative posts.

If you are only working in a batch mode I suggest a steam chest with a flat horizontal surface. Or perhaps just a large heated griddle.

Please tell us the nature of your intended use so that we may better advise you.

[Edited on 23-8-2010 by Magpie]

[Edited on 23-8-2010 by Magpie]

gsd - 23-8-2010 at 17:09

Quote: Originally posted by Magpie  

If you are only working in a batch mode I suggest a steam chest with a flat horizontal surface. ........

Or better still, you can have a flat INCLINED surface heated from inner side and provided with liquid spray arrangement from the top on the outer side. A mechanical wiper type blade can be designed to scrape the product off periodically.

With little ingenuity this this can be made to work as a batch dryer.


denatured - 24-8-2010 at 13:05

Thank you all for your help!

You are right, I don't have to use a drum dryer to dry my proteins. so I think I will go with what you described gsd, I will leave the mechanical scraping for now and just do the scrapping manually. it should be inclined so that droplets gets spread, right? how much inclination do you suggest? how to make the liquid sprayer arrangement? seems like I will need a small compressor and a nozzle (perforated pipe?) and a pump? and some piping. If what I'm saying is correct then I need some drawings to show them to the guy at the metalworking workshop. is there a method to calculate the area that the sprayer will cover with the protein solution (so that I can decide the dimensions of the heated surface)?

Here is an ugly drawing:
ugly-sketch.png - 7kB

gsd - 25-8-2010 at 07:46

What is the capacity you are looking at? 500 gm - 5lkg - 50 kg?

I would advise you to buy a small lab oven and modify it to accommodate few trays and use is as a batch tray dryer.

Within few batches you will get a hang of the process and will get remarkably consistent results. Once you know the time required for drying, it is practically a foolproof process.

After you are done with your drying project, a lab oven will always be a very useful equipment to have.


denatured - 25-8-2010 at 10:17


What is the capacity you are looking at? 500 gm - 5lkg - 50 kg?

I will have a lot of the protein solution, say not less than 30 liters at a time.

I tried a normal lab oven but it resulted in a paste with high moisture content no where near dry powder (I didn't do it myself, I gave the solution to someone who did it).

densest - 25-8-2010 at 12:56

Denatured? ahhh... do you want your proteins cooked or raw? Because a dryer drum will cook proteins very well :D

denatured - 25-8-2010 at 16:10

densest: it doesn't matter, it is going to be used as food/supplement, so some loss in quality is acceptable (drum dried milk wasn't so bad AFAIK). I'm not considering fancy methods right now.

densest - 25-8-2010 at 22:04

For an improvised mechanism, perhaps a spray/vacuum dryer? Spray warm aqueous solution/suspension into a warm dry air current, pass through a cyclone separator, then a bag filter, then a vacuum. Tuned correctly, 90+% of the solids would wind up in the cyclone output and the rest would be deposited in the bag filter.

For testing, use a garden tank sprayer & a wood shop dust collector. Dust collectors are made of ducts, filters, vacuum, etc.... An air compressor piped into the sprayer would avoid hand pumping. A warming belt around the tank would keep it warm enough. An electric heater/fan would supply warm air.

You would be combining stuff easily available from catalogs. In the US, Grizzly Industrial sells this: which may be too big.
Obviously this equipment is not food grade, but that's after the proof of concept.

What you're looking for is similar to whatever is used to make dried milk powder or malt extract powder. I don't think they use drum dryers.

denatured - 28-8-2010 at 00:33

Thanks densest for the ideas but as I said, I'm after something cheap now.

Your idea of garden sprayer is very good so if it works then there would be no need to design an atomizer (for the simple drying method above).

It turns out that I need a scraping knife mounted on the hot surface even if it is operated manually because when I tried to dry some milk using a perfume sprayer and hot stainless steel tray, it was difficult to remove the dried material using a normal putty knife (like this). I don't know how to design that scraper, I'm thinking of something like rails so the knife could slide on it and a method to prevent the knife from going up and over the dried film. any drawings are welcome!

watson.fawkes - 28-8-2010 at 05:18

Quote: Originally posted by denatured  
I don't know how to design that scraper, I'm thinking of something like rails so the knife could slide on it and a method to prevent the knife from going up and over the dried film.
I would suggest resigning yourself to the sure eventuality that there will be, at any given time, a certain thickness of product that's not going to come off the surface. Consider it just a cost of operation.

I should also mention is that the words "cheap" and "custom metal work", as with a special sliding knife for you, are just not compatible.

Magpie - 28-8-2010 at 07:00

Yes, Watson. "Cheap" could include a used Teflon coated electric griddle obtained at a yard sale or Goodwill.

denatured - 29-8-2010 at 09:42

Well, the stainless steel box * would cost me around 100 USD. what do you think of this price?

* = 50 cm x 50 cm, 10 cm height, 1 ml thickness, with holes for the heater and oil inlet, no special sliding knife, he said that would cost a lot.

anandsork - 9-1-2011 at 07:22

I accidently came to this forum & found that denatured had an idea which I was thinking too. I am trying to make homemade spray drying equipment. I don't want cyclone which is hard. So I had same idea. Following are few points which I noted down while browsing internet
1. Some food has more sugar & those are hard to dry. They become sticky when dried. e.g. Mango. The sugar does not crystallize, but melt. That is why the problem. When those are used over drum, they stick to drum & thus reduce efficiency. I think denatured encountered same problem while spraying the material over hot plate.
2. While spray drying, the air enters with ~200 deg Celsius.
3. Spray drying uses hot air stream. I was thinking what if I use induction coil instead?

I am interested in what is the outcome of denatured's experiment.

Doog - 11-3-2011 at 15:51

Wow. I thought I was the only person on the face of the earth that was trying to make a homemade spray dryer. I would like to use mine to dry wheatgrass juice, and other herbal extracts for personal use. The only thing I can visualize in my mind is a large, tall drum. From the bottom, maybe have three hair dryers fed at upward angles to create a cyclone, and also provide warm or hot air. I was also thinking that maybe a Dyson vacuum (the one that uses cyclone technology) to generate the air flow. But three cheap hair dryers would be more economical and also create the cyclone effect. Then I would spray the liquid from the top down. In my fantasy, the dried product would accumulate at the bottom in a powder. An exhaust outlet on top would allow the moist, warm air to escape. That is my fantasy anyway. ;-)

What I am having a hard time figuring out is what to build the container out of. I could make it out of PVC pipe, and line it with greenhouse plastic, as I have an abundant supply of both of those things. But I think the device would have to be quite tall to give the sprayed liquid time to dry before hitting the bottom. Perhaps the cyclone upward air flow will keep the particles suspended long enough to dry before accumulating in a soggy slurry. Maybe I would have to inject the liquid directly into the cyclone flow to dry it before it sticks to anything.

Just thinking out loud. What do you guys think?

[Edited on 11-3-2011 by Doog]

[Edited on 11-3-2011 by Doog]

Doog - 11-3-2011 at 17:19

Attachment: Spray Dryer.vsd (179kB)
This file has been downloaded 717 times

Tried to attach a schematic- but didn't work.

[Edited on 12-3-2011 by Doog]

bbartlog - 12-3-2011 at 08:12

The air coming out of a hair dryer is fairly hot, probably too hot for PVC to be a good choice for any extended operations.
I can't read your attachments as my computer (Mac) does not open .vsd files by default, but in general: what you're trying to do (air drying droplets to create dry prills) is a pretty difficult task, in terms of small scale industrial design. Makers of detergents use (or used, anyway; no idea of the modern state of the art) tall drying towers. This simplifies things. You have a general problem where you need a certain dwell time in order for drying to take place; this varies with the mass of your droplet, maybe even approximately linearly as a first approximation. But if you make the droplets small enough to be dried in a reasonable amount of time (seconds, if your device is sized for your home), then air flow will tend to carry them away, unless it is very slow and laminar.
Your notion of constructing a cyclone is an interesting way to try to solve this contradiction. But I don't think it will work. Generally, these are devices for ejecting dry particles from an air vortex (to settle below / elsewhere). If used in an attempt to dry droplets, I would expect the result to be a wet sheen on the walls of the drum as the droplets hit it.
Anyway, I would suggest working first on a good way to atomize your solution (compressed air blasting across a small feed aperture might work). A fine aerosol makes the drying problem easier, though I suppose collecting the particles becomes harder...