AL811H airflow measurements

811A Tube History and Construction

AL811 and AL811H Cooling System Mods

Unless you change to 572B tubes, there is absolutely no reason at all to modify the cooling system. The stock fan and air system in the AL811 and AL811H maintain components well below rated temperatures, so there is absolutely NO tube or component life improvement with more airflow. The 811 tubes are actually rated for convection cooling, but the AL811 forces air across the tube envelopes anyway. The absolute maximum average dissipation in the 811A tubes is 65 watts per tube, or 260 watts total. This, when added to dissipation of other components, is well within the air system capacity for a 15C temperature rise of internal cabinet air.

Assuming a 15 minute on and 15 minute off cycle, the air system was tested at 350 watts average tube dissipation and still remains safe. If you go much over 350 watts of average tube dissipation for any length of time, you will ruin the 811A tubes in fairly short order. No amount of air, even a gale force wind, will extend the life of the 811A tubes. This is because the anode is NOT exposed to the airflow, it is inside a vacuum, so the extra air does no good. The most extra air will do is cool things outside the tube, and that is not an issue until average dissipation gets well beyond the 811A rating. The amplifier was retested, this time using a 5 minute on and 5 minute off test.

Users should increase airflow if extending duty cycle with 572B tubes substituting for 811A tube. For higher duty cycle operation with 572B tubes, a doubling of CFM is required. The replacement's CFM rating above .08 inches of backpressure must be compared to the AL811's rated flow at 18 CFM. Ideally any new fan would flow 35 CFM at about 0.15 inches water. Some reduction in backpressure will occur by removing the area where exhaust holes are, and replacing it with a 50% opening copper mesh screen. DO NOT extend the cabinet hole area, because the exhaust hole area was chosen carefully to prevent stagnant internal airflow areas .

The original fan is a Pabst 8800N found at this link:

http://www.ebmpapst.us/allpdfs/8000N.PDF

The AL811 and 811H use a fan that follows curve 5 below at about .07 inches backpressure:

Years ago, when I designed the AL-811 and Al-811H amplifiers, I made a series of careful temperature measurements. In light of Internet articles and comments about "improving airflow" that often include adding a larger fan, I decided to re-visit airflow with a current production amplifier and see if anything changed.

Here are internal air temperature measurements of a current, as of early 2011, AL811H amplifier with absolutely STOCK cooling system:

50% duty cycle carrier, 450 watts output, 5 minutes on, 5 minutes off

At that point the tube anodes are red, which is really hammering the tubes.

Lock-down idle, 2 hours, no RF just quiescent current causing 170 watts total tube dissipation (this result will vary with idle current)

Extending Tube and Component Life through Increased Airflow

This sounds nice, but unless we are melting the tube's glass or ruing seals any additional airflow is unnecessary. The 811 tube anode or plate, which is the weakest element in the tube, gets rid of heat mostly through infrared radiation. Infrared radiation is a form of electromagnetic radiation, but well above radio frequencies. Very little heat is conducted out of the tube. The anode cooling is primarily determined by the temperature differential between the tube anodes and the surrounding objects in the amplifier. it is heat radiated out for a large distance, not conducted to the glass.

In the 811 tube, normal unforced convection airflow in a large box can adequately maintain tube glass and seals far below rated temperatures! With 811A tubes in the AL811H or AL811, forced airflow is better than already adequate convection in a large box. Actually, this is one of the reasons the cabinet is so roomy. With a tighter cabinet, more airflow (and more noise) would be required for the same cooling. Modifications to the stock air system cannot increase tube or component life. In this case tubes are dissipation-time limited by characteristics inside the tube. Tube life is not even remotely close to being influenced by cooling system airflow volume.

Now in fairness there is one thing that might require more airflow. The  AL811H, as well as the three-tube AL811, might need more airflow with 572B tubes if the amplifier is operated significantly harder than with stock 811 tubes. This should not be done anyway, however, because the power transformer will become unreliable with significantly higher duty cycles or output power. 572B tubes, substituted into the 811 series, should not be driven significantly harder than original 811 tubes. At approximately 800 watts PEP voice IMD products are -32 dB PEP. Beyond 800 watts IMD will be worse than most radios, and considered excessive. This is because of filament emission limits in the 811 and 572B tubes. Both tubes share the same filament designs and have the same peak emission. The 572B primarily buys extra time or extra anode dissipation margin for mistuning safety or use of heavy speech processing, not significantly more peak power.

The AL811 amplifier, with the fresh look in early 2011 to look for any possible improvements, is actually a very good design. All components are sized conservatively enough to not be a problem except the tubes themselves, but not wastefully oversized. While it would be nice to boost sales and get more life into the product line (and get paid to do it) by improving the 811H, there really isn't anything worthwhile that can be done. Changing any single part or small group of parts, short of making an entirely new amp, would really be a waste of time and money.

Where did the idea the cooling system needs improvement come from?

Perhaps a little of it came from another 811 amp. Years ago, someone else marketed a four-tube 811 amp. They used a larger cooling fan than the AL811, and openly promoted this as a "feature". Perhaps Hams just equate life increase to airflow increases, as with hot running semiconductors or hot running electrolytic capacitors, so they think more noise and fan size means more life. Unfortunately the amplifier, with much more noise and fan speed, actually ran much hotter internally than the 811 series! While a larger fan was used, the fan inlet was jammed up against air restrictions.

Restricting the inlet or placing things near the blades does two very bad things to almost any fan or blower. First, air moving through or following the blades is disrupted. Noise increases from this, and airflow decreases. Any inlet restriction is much worse than an outlet restriction. Restricting the inlet has a much greater effect on airflow than any other restriction, and restrictions generally increase fan speed and fan noise because the fan cannot push air smoothly with minimal turbulence. The fan really needs to be in a clear area, particularly on the low pressure inlet side.

The second problem was the manufacturer used a very compact cabinet without much room around components. While that is great for desktop room, it makes the amplifier much more dependent on forced-air. The AL811 series was primarily designed to be inexpensive, as reliable as possible for the tubes used, and quiet.

Perhaps some backyard engineer somewhere decided his melted anodes would be corrected by blowing more air past the tubes. This would be true if the anodes touched the airflow, like in external anode tubes, but it is not true in the 811 tube at all. If the glass is maintained significantly below damaging temperatures, and if cabinet internals are not significantly above inlet air temperatures, additional cooling is meaningless.

Tube life is almost exclusively determined by duty cycle and tube quality. Other than keeping tuning times short and avoiding heavy speech processing or duty cycles that actually melt anodes (overheated anodes appear as a silver color or shiny area on the anode), there is little that can be done to improve tube life.

I am doing a page on new AL 811H changes. That page will be up soon at this link.