The AM Forum

I wonder if anyone has experience stacking two identical muffin fans coaxially. (one on top of the other)  The fans are standard 120V  5" X 5"? types you comonly see used for cabinet ventilation. Does the technique work?

I have a limited footprint space for just one fan and desire to increase the flow with a slower speed. Should I stack them with the air flowing in the same direction? I was once told by an enginner to mount them back-to-back in different directions. This seems odd, but that is how he did his amplifier.

Or maybe it's better to use just one fan. The main objective is quiet operation followed by air flow.  I will use a Variac to ramp it down.

I see little on the web about it.

T

Mock up an experiment. Id suspect there will be a bit of turbulance to contend with especially if the rotors arent phased exactly. Ive seen extra thick muffins and that may be a better choice.

Carl

JJ,

Is there anyway you can use them in separate locations, one intake and one exhaust?

One of the central vacuum cleaner companies uses fans in series to develop greater suction but I doubt this is of much value unless you have significant airflow resistance.  It turns out the series fans have a fairly short life since the second one sees high air temperatures but this won't apply to your scenario.

I have found that fans of the same size vary greatly in their ability to move air along with quietness.  When one of the pedestal ventilation fans went up in smoke on my Desk KW I replaced both pedestal fans with a pair of Rotron patriot series fans.  They run in series across 120 volts with far less noise and greater airflow than the original fans.  I assume this is largely due to the more intricate shape of the plastic blades compared to the early simple aluminum fans.  You probably can find a single fan that would move more air than a stack of two.

homebrew a manometer.

The manometer is a good idea, a few feet of clear tubing and water, form it into a u-shape and attach it to a board for rigidity, add a couple drops of food coloring so you can really see the level. you can then convert inches of water into flow rates if you know the orifice size.

I don't recall exactly, but I seem to remember that stacking fans didn't work out so well.  The flow would stall between the two fans and efficiency was not close to 2x but a lot less.

2 fans side by side would be better methinks.
My fluid dynamics course was some time ago... ;)

I think just stacking one on top of the other would be ineffective.  When one fan runs at a given speed, it moves a certain volume of air per unit time.  The second one, assuming a stationary frame, would see that volume of air already in motion at its intake, so its blades would spin at the same rpm but loaf along, doing little work beyond overcoming its own friction and turbulence. If you cut power to the 2nd one, it would continue to rotate close to normal speed, its blades blown like a windmill by the first one.  It would seem to me that the 2nd one would have to run at a higher RPM than the first, in order to move significantly more air, but then it would try to pull air through the first one and speed it up so you would still have the same situation in reverse. You would need to mount them side by side or at different localities in the enclosure so that each one would move its own volume of air.

Where stacking two fans might be helpful would be if the motor in each fan were too small for the size of the blades, and the blades were loading down the motor enough to slow it down excessively.  Each fan would help the other along so that they would both run at normal speed.

Stacking fans would be analogous to what I have observed happening in my two-tube final.  The filament goes out on one tube and the other one tries to take the full load of the two.  If it is a really good tube, capable of running twice the normal plate current, grid current and plate dissipation, you might not notice anything abnormal from the meter readings if you couldn't see one cold plate and the other one running white hot.  The only way to make the final operate at half power with one tube turned off and the other operating normally, would be to make adjustments to reduce the antenna loading and grid drive.

Yes, you can series fans but you need a cylinder section between the two with straightening vanes, actually a set of bundled smaller tubes inserted inside the cylinder. I don't remember the exact formula but I think six diameters diminishes the turbubance enough.  As previously mentioned your better off with push pull.  ...one on intake and one on discharge of the cabinet.
..unless you like the Queen Mary look, urk.

I've seen the stacked fan thing in 1U rackmount servers. Very noisy. I suspect they do it to increase air pressure, since the fans are so small and the servers put out so much heat in a small, restricted space.

A better option, like others have mentioned, is to do an intake/exhaust combo. To make it more silent, use big, high-quality 120mm fans. More air, lower RPMs, less noise.

2 seperate fans, side by side, will move 2 X the air

2 stacked ( as already stated) will NOT move 2X the air.

The second would need to be rotating faster than the first. ( Assuming identical size, rotor size and rotor angle)
Even then the efficiency would be not good.

2 fans, one intake and one exaust, will move a little  more air than 1 fan.
(depends on the framework of the fan body)


Best scenario using 2 fans is having them side by side... blowing the same direction.
Obviously there must be adequate static vents to cope with the flow.
( surface area of the vent openings should be at least 1.5 or 2 times the surface area of the fan openings combined)

Counter-rotate them to gain any efficiency with a bit of separation between them, otherwise install the side-by-side.

Thanks for the suggestions, guys!

Interesting on the counter-rotation.  That's just what Chuck/K1KW did with his two fans in his linear amp. His were about 10" X 10" and opposed. He said that was the way to do it, but it seems counter logical. IE, how does the air "know" which way to go?   ;D   Evidently the path with the back pressure is the way.

I should have mentioned before that this application will have some back pressure through a pair of 4CX-250's ext anode tubes. This is the next stage for the HPSDR 10 mW rig. Class A modified TMC amp I've been working on in another thread.

Some background:
As seen in the pics below, I can mount a muffin at the bottom of the enclosed final sub-area and seal it off. The air top would be free flowing thru the tube fins.  The amp originally used a squirrel cage. It was bad and I replaced it with another which was too darn noisy. So I’m trying a muffin fan and hoping it can do the job since the amp will be run only at about 40w pep out in class A. Just built a power supply with only 900V plate, 200V screen (reg) and -30V regulated bias. Tubes: 6CL6 > 6146 > (2) 4CX-250’s all in class A1 with heavy RF negative feedback.  Hopefully -45db 3rd IMD and super clean at 40W pep out.


I don't think I will need an exhaust fan on top.  Hope the muffin can handle the back pressure. Think I'll make a cardboard mount and try it first... ;)  Going to try the counter rotation too.



T

The back pressure from 4CX250 tubes may well be a problem for a muffin fan.  That was the tradeoff with the external anode tubes and as I recall the 4CX250 increased rating was largely through a larger and more efficient anode cooler but with greater airflow requirements over the 4X150.  Although you aren't asking them to dissipate too much (I am guessing about 200 watts input to allow for typical class A efficiency plus output circuit loss) they will be running at close to 100 watts average dissipation per tube.  Of course the more and louder you talk the cooler they run in class A :)

I see for 200 watts per tube at sea level (more than you need); specs call for around 5 CFM but with over 1/2 inch of drop which is a lot for a muffin fan to provide.  Many of the little fans drop to 0 airflow at around .3 inch or less of pressure.  I have a feeling a fan that will overcome the back pressure is likely to be high speed and fairly noisy.

I would take a close look at the specs for the fans you are thinking of using.

Buy some of those temperature sensitive sticks from a welding supply and make marks on the tube base or envelope.  The key will be if you do not exceed the seal temperature of the tube in question.  Should you notice it is close then you need to increase the airflow.  

Tube specs. tell the story, be sure your installation meets those specs.  Or you can order from a mail order place (amazon) or go to hardware store and buy a infrared thermometer.  Get one with a distance ratio that fits your needs and check the tube base to read the temperature.  Adjust your airflow to maintain the temperature below the max spec. at @1.5 the specification.  If you don't have the horizontal space, then the stacking of the fans is the way to go provided you have proper space between the frames.  The greater the airflow of the top one the wider the spacing.

Rodger,

Yes, you are correct.  I just ran a test using the muffin fan sealed off with a cardboard support. The back pressure killed the air flow to a tiny trickle thru the tube fins.  Looks like I will need to go back to a squirrel cage, darn.  I have some smaller ones I'll try with a Variac.

I had a pair of 6146's in there just yesterday and they needed no air. But they were good for maybe 15-20W out class A, so put the 4CX-250's back in.

Another problem with the squirrel cage is it is mounted down below and the amp will need tall feet to let it clear. 

OK on the temp stick-ons, Jim. Yes, I'm real careful with cooling these days after losing a few tubes to heat abuse in the past. Now I use outside air sucked in that is usually cooler than room temp.


I'll post the solution, whatever I come up with.

T

I actually performed a test like this back in the 80s when I was a 'JN, blowing up things one day (it didn't start out that way, honest). My dim recollection was that stacking two or more created some kind of Venturi effect with one fan suddenly ramping up considerably from whir-r-r-r-r-r to weeeeEEEEEEE. Probably related to what Don mentioned.

Muffin fans were a great source of entertainment to me back then (it didn't take a lot). Had several different sizes. Discovered that they do break, including the blades, if you play too rough with them. And that fine wire wound inside, well.....let's just say they weren't intended for high speed ducted fan service.   ;)

Wow, the tube outlet total surface area (tube fins) is VERY small. I'm finding even a big squirrel cage pushes thru very little air.   Highly restrictive.

Gonna try the back-to-back muffins next.

T

Since space in the amp proper is an issue, how about remote mounting the blower and connecting with air duct?  I did this with my homebrew triple 4CX800 setup.  The blower is mounted in a large wooden box with insulation on the inside surfaces which kills a lot of the noise.  The intake is through an air filter so dust buildup in the cooling fins is greatly lessened.

You can build one nice blower setup and use either quick connect air couplers and or air valves (sold for use in wood shop dust collection systems) and use your blower system with multiple amps.

Yep, Rodger, I've done the remote blower thing before. All of my big rigs use them now.

Well, the counter-rotation muffins were a bust. It made X3 the noise and didn't blow much air.

The two in-series blowing in the same direction seemed about 1.5X better, but was twice as noisy.  One alone was not enuff air thru.

So looks like a blower is the only way.

T

Tom a good muffin fan will give you about 1/8 inch on a U tube.

T- glad to see you're puttin that pretty little TMC amp to good use!

Didn't that thing already have a fan on it?  Or is it too noisy?

Yo Johnny -

The fan was open when I got it. I put another one in it and used it for a while.  It's been thru a lot of changes, but in short, built a new power supply, installed full power switching and antenna switching relays, as well as fixed a few things. But it gonna have good service as LeeLoo's linear.  I even changed it over to two 6146's in the fibal for a week, but in class A it was not enuff power to drive Dr. Love. So back to the 4CX-250's in low voltage mode.. (900V)

I added some more NFB and might add even more since I have plenty of drive and gain to work with.

They sure are built well.  Did you ever use the log periodic I traded you for it or is it sitting in a shed somewhere?  (with the wasp nest)  ;)

T

Do we have an aeronautical engineer type on board?  I was thinking of those Russki bombers with all the counter-rotating props.  They seem to fly OK.

73DG

They make a centrifugal blower that is the same height as a single "muffin fan", cheap in the surplus places, that will work into your backpressure type load. The output is on the side, but so what... it will push around a corner ok... you could
mount two of them if there is enough room - I think they also make 3" versions, so maybe you could place 2 or 4... slowing them down is not hard, easier if they are DC motors... I'd check the usual surplus outfits to start...

                    _-_-bear

My experience agrees against stacking them directly. Plus the noise factor is there whether it sounds cool spinning up or not. With or without vanes, separate them. One pushing and one pulling, at opposite ends of the overall flow path is acceptable. If you need more volume, put them in parallel. If more pressure, a single larger or higher speed fan or maybe even a blower would be better.

Bear, do you know the brand of that fan you mention? I think I know what you are describing but seldom see them.

Why dontcha just place 'em in water.... ..


klc

If you had two enclosed fans opposing (facing each other) in a "T" configuration, the outlet would certainly have more pressure and higher flow. Maybe this is what he meant.

By a "T" arrangement I mean two funs in a duct pointing at each other with a third duct at right angles to each fans exhaust. This third duct could be the one pressurinzing the tube bases in the RF deck.

Phil - AC0OB

The two fans into a "T" sounds pretty promising to me, but I think some kind of ducting that would eliminate the fans blowing straight at each other, leaving the combined airflow trying to figure out where to go would be even better. Visualize the difference between a regular plumbing "T" and a "sanitary T" for wastewater. I think the sanitary T arrangement would yield better airflow by minimizing the turbulence as the outputs of the two fans meet and helping to blend the two airstreams.

ldb
K5WLF

RF Parts sells a centrifigal blower for the 4CX250 tube.  It's their replacement for the Pride DX300.  It was a WELL abused amplifier, you may find it will move enough air for both your tubes....  I've seen a few of the single tube amps converted to dual tubes.....  And most kept the original blower.

Just a thought, it's pretty small.  And, as Bear stated, their is a 5 or 6 inch blower the size of a muffin fan....  Can't remember where I saw them, the bookmarks are on another (non-connected now) pc.

--Shane
KD6VXI

I have the answer, it does work on 120 volt AC, 5 inch fans. My son does it to cool LiPo batteries that he's using in his RC car. He's into RC racing  like we're into ham radio and has taught me a lot about charger controllers etc.

So anyhow I gave him two identical fans and he stacked them together and I have witnessed the affects of a much stronger air flow than just using one. No just stronger but impressive!

I still have six left and I'm going t stack two to cool an 833A at the camp. Right now I have a 220 volt fan running on 110 to quiet it down. It a bear on 220.

I don't know how this may work with DC fans with the permanent magnets in them? I suspect with the AC fans there is some self synchronization going on where as with DC types, there may be little slippage to have the (theoretical+) self synchronization going on.

I'm going to mount three more of them in a hard hat so I can cool my head from three sides.  One will go under this chair I'm setting on the cut down on BS :P :P :P

+ based on no facts.

Tom, haven't done anything with that log yet.  I'm waiting for the right moment  ::)

This had some interesting comments about that.

http://en.wikipedia.org/wiki/Contra-rotating_propellers

I wonder if it's not too applicable to the problem at hand because of the difference in Reynolds numbers?

The counter rotating props still BLOW in the same direction, yes??
Opposed stacked muffin fans do not...

centrifugal blowers:

http://www.allelectronics.com/make-a-store/item/CF-371/12VDC-50MM-BLOWER/1.html

http://www.allelectronics.com/make-a-store/category/220300/Fans/Blowers/1.html

Marlin P Jones might have some of these as well...

If is is flowing and not allowed to enter vapor phase that might work. The fin spacing is so small that bubbles could become entrapped in the radiator.

It should be pointed out that noise is usually generated by turbulence and the generation of turbulence represents a loss of energy. When building up ducts and size transitions for airflow, it is best to fill corners with putty like 'DAP' and also create contours leading into and exiting any restrictions, even small ones. Case in point is an A/C evaporating unit I built for my previous M35 communications truck.

It was esssentially a box with a coffee-can-sized hole in one side for the inlet and the A/C evaporator making up one wall as the outlet. Inside was a 450CFM centrifugal blower from which I had removed about half the scroll. The remaining half supported the inlet airfoil and tube. It was just 1" diameter short of the same diameter as a large coffee can. Two such cans were used to bring outside air into the box and to the blower inlet. It was well-sealed. As-built, the noise from the intake was horrendous. The noise annoyance was reduced by 2/3 by building up a shape in the transition between the inlet and the coffee can diameter, getting rid of the square profile of the transition.

Take a look at EBM-Papst fans. High quality, all metal, and tight tolerances. The LK-800 even with a full bore 5KW out mod uses one to move loads of air thru three 3CPX800A7's. At lower power a 330 Ohm series resistor is used. About $27 at Gopher Electronics but it may be too wide, check out their smaller ones on line.

Carl

Thanks for the suggestions, guys.

I ended up going back to a standard blower on a Variac. Seems the 4CX-250's have so much restriction that they aren't designed to flow lots of air like a big ext anode tube. I wasn't used to this.

But in the end, a muffin fan, in any configuration, is not as good as a standard squirrel cage blower when dealing with back pressure.

T

Tom,
       Kevin might just be onto something....................

I have seen 4CW250's pretty cheap at hamfests. They are exactly the same specs as a 4CX250, but have a water cooled anode. they were once popular for high powered VHF / UHF stuff but people got tired of futzing with the cooling plumbing. With a water cooled anode, you could really beat the snot out of them!! (and no fan noise!!)

Just a thought..............................

I've seen liquid cooling used on a 4cx250 .... used hv oil and no pump ....cooled by convection .... specific heat MUCH greater for liquids rather than air ... distilled water ok but must be changed fairly often due to ion buildup ... biggest problem is how to avoid leaks ...73...John