The AM Forum

I have decided to add a pair of small DC brushless 24 V fans to a homebrew 811A amp. I was planning on NO FANs. However, the compact size forced me to rethink about getting the heat out. Unfortunate, these fans don't like 200 watts floating around, shuts them down! :( I can try RF bypass on the FAN supply line, chokes and beads, but perhaps the best approach is go with a 120 V AC muffin fan.

Thoughts? Comments?

Thanks! Alan 

I like simple solutions .... one I just learned from Cactus K4YMB is to put a 1 uF oil filled cap in series with the 120V feed to a muffin or other shaded pole motor fan .... cuts the speed in half and all runs cool ....ymmv

Thanks John. I'll give that a go. Not sure in a high RF setting, without effort to decouple and shield, that these low voltage DC brushless motors can tolerate the RF voltage present. They apparently de sense and shut down.

Where are the fans mounted? If they are on the chassis via a vent hole, a fine wire screen between the fan blades and the chassis might do the job?

If series feeding an AC synchronous fan via capacitor watch for startup speed if any. Might be too large a cap ;)

Good point. I will try. Currently, they are in a poor location RF wise, within the RF cage!! Obviously, they were not planned into the scheme.

Maybe placing them elsewhere where they still suck in air, and using a hose or made up duct of some sort might help. If you have the space.

Thanks to all for the replies and inputs.

Placing the FANs outside the cage, directly above the plate caps and blowing air down at the tubes, fixes the issue. And, I had to add 0.15 uF cap directly across the motor DC terminals. I still need to button up the sides of the amp, that is add left and right walls. I'll probably try to use cane metal for ventilation reasons, unless it's crazy expensive. For now, just top cover is cane metal and that seems to work well.

Thanks again forum!   
Alan

Hi Alan,
Glad to hear you are making progress with your home brew project. One comment and a question.

As a rule of thumb, you might want to use your fan to remove heat from the tube chamber and not blow air on the tube. This idea holds for transistor heat sink's as well. Make sure you have vent holes around the tube socket and suck air thru the vents, over the tube and exit thru the fan. Blowing air at a tube can cause uneven surface cooling and envelope expansion/shrinkage causing possible seal failure.

My question, what is cane metal? Never heard of it and a search doesn't turn up anything.

Mike

Hi Mike. Thanks for the note. I agree, I do not like the arrangement I currently have and it's a short term solution for sure. I need a better layout arrangement and looking at that now. The 811's are mounted below the surface of the chassis so their is a "ring" around the tube base that air could be moved through. Again, this is short term and on the cane metal...

Just perforated aluminum, punched with periodic vent holes. Works well. A photo below. I should have called it "perforated aluminum", but in the past I have referenced it as cane panel. 

If you evacuate hot air through the fans, the motors may get hot unless you are moving so much air that the cane metal stays cool. It is possible to do that but some times noisy.

Most fans, but not all, are better at blowing air through a resistance and it is usually easier to blow air than evacuate it through cane metal when the fan is right against it. Depends on the shape of the blades and their closeness to the obstruction.

Under the tubes is usually the best place for air inlets but it makes the desktop chassis taller and offends aesthetics for some people. Thus all kinds of compromises like the Heathkit 3-500Z cooling, the too-small blower under the NCL-2000 chassis, etc.

Fans do not make much pressure so there may need be many chassis holes to get enough flow to cool as well as you are now.

If the tube compartment has solid sides and cane metal top and many holes on the bottom, then there can be much air flow forced over the tubes and less wasted escaping from where it's wanted

There is nothing really wrong with 'reverse' cooling if everything is properly cooled.

Maybe the real point is to experiment and make sure all parts are within limits when you beat on the amplifier. One of those IR thermometers or thermocouple might help.

If RF interferes with readings, you can temporarily bias the 811's positive, respecting grid dissipation, and run the plate dissipation to the proper amount and check. (+4 to 6V should be enough at 1KV to get 50-60W plate dissipation) Done that trick many times on 4-1000s and 3-500Zs. Not orthodox I guess.

I always saw 'cane metal' as having the more ornate style of perforations but it is in the eye of the beholder and beauty is wasted on fans. I also mix the term sometmes.

I have used those DC brushless fans before,  in both solid state and vacuum tube gear.

Solid State to a legal limit plus mobile amp,  16 x 2sc2879s 40 to ten meters (no antenna for 80).

They will shut down,  chatter,  act as a modulation monitor,  etc unless fed CLEAN DC power.

Ferrite beads,  with many turns,  at the fan.   As close as possible to the fan. . Judicious use of. 1,. 01 uF disc caps, and leads as short as possible.

I've seen 6 inch brushless fans cooling a 64 transistor amp.  It had north of ten grand of rf in the chassis:  I cloned my bypassing method from that one.


I used the same method with 2 small muffins on top of my FL2100.  It wouldn't stop,  but it would slow the fans down.   They heated up.

Also,  from my mobile amp days.   The closer the obstruction to air flow,  the louder and less efficient the fan is.

The fans on my 12 volt power supply are standoff mounted to the rear heat sink.   Mounted directly to the fins,  I halfed the airflow,  and over doubled the noise.   Mounted on 1.5 inch (iirc)  standoffs,  nice,  cool and more air is moved.

--Shane
KD6VXI

Again, thanks in advance for the valuable advice.

Here is a top view. As I see it, the only options are rear panel sort of centered on the tubes, that is aiming air flow between. Or on the left side. The cathode match enclosure and the shielded meter casing distort the air flow on the left. The rear is not perfect either, however, I think the distortion is a little less. A single fan is 4.5 inches on a side. Not sure of the CFM, although I could probably find the data sheet. A chassis bottom mount fan is really not an option at this point.

Alan 

Top left rear corner I'd slap a hole there and have a fan pushing air in or pulling air out.

Not ideal,  but it would provide steady airflow.   Nothing I  the near stream to upset flow either.

Convection cooling doesn't take a heaping helping of airflow usually.   A decent amount of movement of air unless your beating hell out of the active devices.

--Shane
KD6VXI

Hi Shane. I agree. It's about the only decent area remaining. It's hard to get it perfect when it becomes an after thought!

Thank you for that practical advice especially on the large DC powered stuff!

That's almost too pretty to make any holes in! Beautiful!