PCM Range Testing

FAIRCHILD PT-19

Many, if not most, intermediate-to-advanced modellers have developed some system for checking their battery condition. And rightly so. Batteries, from what I can tell from my 30 years of flying, are still the number one reason guys experience catastrophic failures.

In solving the battery dilemma, most fellows know a numeric value of their battery capacity through use of a cycler. For example, if I ask a guy what his battery is, often times he'll say "I have a JR Extra 1800, and it checked out at 1790 last time I cycled."

Perfect. He's taken the time to get a handle on exactly what his equipment is doing - a great way to save having to re-build a model. Then, if checked before every flight with a battery tester, the chances of having a problem are virtually nil.

Oddly enough, if I ask the same guy how his range checked out, he'll say "okay - fine - pretty good" ... all
qualitative answers. The reason why is clear- there exists no standard system for how to quantify range.
This article is an effort to develop such a system. While most of the concepts are directly applicable to FM, we're going to utilize the "Fail-Safe" system that's exclusive to PCM users to help establish a criteria for measurement.

"Range testing is every bit as vital as battery conditioning/cycling.
This article gives you a system to follow to treat range testing
in the way you treat your batteries."

Comprehensive Testing

FACTORS INCLUDED
IN A RANGE TEST

USING THE SYSTEM DESCRIBED

·model generated rf and emi

(bearings, pumps, batteries, loose bolts, mufflers, etc.)

·transmittter output

(module condition)

·transmitter frequency tuning

·receiver tuning

·receiver sensitivity

·vibration test

·receiver antenna placement

·model structure interference

(receiver signal re-radiation)

·transmitter/receiver crystal condition

Range testing is comprehensive in nature

Unlike battery testing, which is focused on one issue - battery capacity, the range test is very comprehensive, taking into account a host of different aspects critical to the success of your model. There's no doubt that many modellers are successful without performing a range test. But having performed some "post mortem" reviews of models that met their demise, it's clear that many may have been saved had the modeller performed an adequate test.

While it's nice that many modellers have confidence that their JR equipment "has worked flawlessly in the past" and therefore conclude it should be true in the future, it just isn't so. As the sophistication of modeling increases, the number of variables that affect radio performance increase. JR builds a superb system, but even the best receiver in the world isn't going to be able to reject the interference caused by a 35,000 volt spark ignition system that's placed in close proximity to the receiver. That's a gross example, but there are many other examples of model-generated RF, the cumulative effect of which is to degrade radio range. In polling JR team members, all agreed that model-generated interference is equal to or more common than product-related or extraneous interference. The ONLY way to test for model generated interference is through an engine-on range test. Clearly, the more expensive the model, the greater the need for a comprehensive range test. What's more, AMA safety rules wisely call for every model to be range tested. Let's get on with it.

Why PCM

Why the range test is an important indicator with PCM

These days, top-level modellers use PCM. Frankly, I can't think of one who doesn't. The reason is simple - PCM provides superior control for our sophisticated models. FM is fine for sport models, but as sophistication grows, so does the need for PCM. While some mature modellers are still leery of PCM- due to lock-out experiences with other brands 15-20 years ago - today's PCM equipment can't be beat for keeping your model under control. That having been said, it's also true that PCM can, and will, go into "hold" (or fail-safe- a preset user programmed position) if it does not receive the correct coded signal for a given period of time. Symbolically, one could think of going into hold as the same thing as "going over a waterfall". The purpose of the range test is to give us an indication of "how close we are to going over the waterfall". Good ground range indicates that the system is strong - it will take considerable interference from outside sources to cause the model to go into hold. Poor ground range does not mean the model will crash, only that it is more susceptible to outside interference - the smallest interference might cause the system to go into hold. Because models are so tough to come by, personally I prefer robust ground range!

The System.

Normal technique for range testing calls for the operator to wiggle the stick, and a helper to determine whether the system is working or not. Problem is, they aren't connected. The helper doesn't know what inputs are being provided, so determining when the signal is lost is murky. To address this issue, the method suggested here uses the fail safe part of your radio to determine the exact point where the radio loses range. By programming the elevator to kick "up" when the signal is lost, it becomes perfectly clear where the exact point is - allowing us to quantify the results. In a sense, this is a parallel to choosing a .9 volt/cell cutoff point for battery conditioning.

300FT MEASURING TAPE

Bob Sadler holds a 300 foot measuring tape, purchased for about $35 at Home Depot. With it, you can record the exact distances of engine-off and engine-on testing, for quantitave comparison. Any significant reduction would lead one to consider looking for troubles - ignition modules giving off interference, transmitter modules low on output, or out-of-tune receivers are common considerations.

Site Selection

Your flying field is best, but to do this right there must not be any other transmitters on while you're testing. The reason? Their antenna will be up and they'll be putting out high power compared to yours. Will it affect your model? Not necessarily, but possibly. Next best is an area where you can control the variables. The point is to get an area where you can have consistency - year-after-year. You want the system to be repeatable, to look for changes that might be an indicator of degraded performance. (Note: Horizon's R&D manager, John Adams believes the field
is far and away the best choice because it will give you an indication as to whether any site-related interference exists.)

1 Place the model perpendicular

Place your model perpendicular to the direction you'll be walking. Which side faces you should not be an issue (test it if you're concerned).

2 Program Up-Elevator Fail-Safe

Program your radio so that when you lose your signal, the elevator goes full up to provide your helper with an easy
indicator of when you lose the signal (don't worry, we're going to take this out in the next phase!) It's pretty easy with both the 8103 and the 10X systems - after a few tries it'll become second nature. To check that your fail-safe is programmed correctly, simply turn the transmitter off - you'll see the programmed surface move to its commanded position.

3 Collapse or Remove Antenna.

There seems to be a lot of confusion regarding whether one should check range with antennas installed or removed.

Here's the correct method:
JR 10 channel transmitter: Antenna REMOVED
All other JR transmitters: Antenna COLLAPSED

Because of the 10X's unique ball-swivel feature, the transmitter's output without the antenna installed is roughly equal to all other transmitters with their antenna's collapsed.

Worried about hurting the 10X's output? Don't be. It's true that lengthy, continuous operation of a transmitter without the antenna places extra load on the output transistor in the module. Your module will get hot and it is the heat that weakens the transistor. But for short periods - 5 minutes or less, heat build up is minimal and not a factor.

Range Testing

Team Member Mike Gregory shows us good form in the range test. Pat Hartness' field is an ideal site for controlling the variables - an open field, with no planes present.

4 Hold the transmitter at arm's length, facing the model

This helps with consistency - see "Variables" text.

5 Walk backwards until the programmed surface deflects

Once the receiver goes into hold, drop your hat to mark the spot where it went into hold. This is your reference point. Record the feet (or paces). You now have your engine-off value.

6 Re-program Fail-Safe for engine idle

Remove the surface deflection you used in the engine off test and program the radio to go to low throttle (idle) in hold. This is a good habit more and more modellers are following - not only for an indication of when your model is at the limit of its range (without affecting its direction of flight), but for good common sense and liability protection.

7 Perform the exact same test with the engine running at mid-power

Next, start your engine. Have a helper restrain your plane, or use a nylon rope with a sturdy stake to hold your plane in place. Advance the throttle to mid or full throttle and then with the transmitter held in the same manner as the earlier test, walk backwards until the engine drops to idle. Mark the spot again, and record the feet (or paces) heading back to your model.

Your range testing is complete. Whether you re-program your model to remove the throttle fail-safe is your choice. Most professional modellers I know choose "hold" for all functions except throttle - which is pre-set to idle.

Think your restraining system won't affect your range?

Think your restraining system won't affect your range? Think again. When this Advance was restrained by the Midwest Aero-Mate (on the left), we got a reading four times what the range was when restrained by the nylon loop on the right. Evidently, the metal yoke re-radiates the signal for a false reading.

What's Good Range?

" If your engine-on test loses more than 20% range, investigate."

You should now have two sets of numbers. Ideally, they'll match. Most times, especially with spark ignition engines, you'll see a slight reduction in ground range. How much is too much? Here's a rule-of-thumb most fellows are following:

If your engine-on test loses more than 20% range, investigate.

The thinking is that a 10% error could be due to the testing system. However, I've seen some fellows who've had their engine-on test result in 1/3rd the range of their engine-off test - a clear indication that something isn't right.

Click to Enlarge

So the big question is: what is good range, and what's a minimum to fly? Variables are going to contribute greatly to the numbers you've received. I've had some basic sport models, tested on blacktop, with perfectly tuned systems exceed 500 feet of ground range. How low to go is largely a personal matter, reflecting your personal desire to accept risk. In polling most JR team members, internal and external, the rule that most seem comfortable with, measured under "normal conditions"- grass, level, model on ground is this:

Minimums:
Engine off: 175 feet (60 paces)
Engine running: 150 feet (50 paces)

In Conclusion

Range testing is a technique. Like all modeling techniques, it's open to interpretation and adaptation. I've checked this system with some of the top radio guys in the country and they've given it a "thumbs up" for methodology. So before you deviate, you might try it this way first.

If nothing else, hopefully you will have improved awareness of the range testing process. Most JR flyers I've met are interested in high performance, and as a whole are better-than-average modellers. I hope this will help protect your investment.

Variables

VARIABLES THAT AFFECT RESULTS
Besides the basic equipment-related issues that affect the core of range testing (receiver tuning, TX tuning, antenna placement, etc.) there are a significant number of variables which will affect the number you receive - all of which point to using the same location, the same methodology, if you are going to measure performance on a year-to-year basis. Here's a few we're aware of.
Type	Significance	Why
Surface type	Significant	At a local private airport, we tested models on the closely-cropped grass. At the extreme end of the range, we then walked onto the center of the runway - and proceeded to get 20-30% more range consistently. So the surface over which you test your model DOES make a substantial difference. Grass gives shorter numbers than blacktop.
Model height	Significant	Usually, for convenience, we put models on the ground. However, depending on the size and placement of the receiver in the model, we usually see an increase when the model is placed on a non-metallic table. Needless to say, this more accurately reflects the model's actual flight conditions. However, it's important that the table be sturdy enough to handle engine-on testing.
Transmitter height	Somewhat significant	When you reach the limits of range, a transmitter that is low will have less range than one that's held overhead. Holding your arm out horizontally removes a variable.
Surface slope	Significant	We discovered that a sloped surface can affect range. When the transmitter was below the model, we measured less range than when the transmitter was above the model. In the case we tested, our range varied by about 20% simply by swapping position of the TX and model.
Modeller proximity	Potentially significant	We've found that often, the location of a modeller's hand around the fuselage where the receiver antenna is located will increase range falsely. It seems that if modellers are standing around, there is no significant change. But having someone hold your model around the antenna is not optimum.
Transmitter/Receiver voltage	Insignificant	Common sense would dictate that higher voltage would improve range, but it just isn't so. We did a range test with a 10X that was beeping due to low voltage (8.9 volts) and found the range to be virtually identical to an 11 volt transmitter. Voltage of the receiver pack also had no affect on range.

Pump problem

One year, at the Florida Jets, Bob Violett had his brand-new (2nd flight) Bandit go into hold on landing. Fortunately, he got it down okay, despite intermittent control. He asked me to check it out. The engine-off test showed that he had about 50 paces of range- not bad. Engine on, he had only about 15. The culprit? I noticed the antenna was in close proximity to the turbine's fuel pump (about an inch away.) By re-routing the antenna to the other side of the fuse, we tripled his ground range- and he has put over 200 successful flights on the Bandit. My guess is that while his test flight in private was fine, when landing with plenty of transmitters close by, the radio went "over the waterfall".

Lucky save

A fellow giant scale flyer, with a spark ignition system, was noticing that his model was occasionally "not locked in". We did a range test with his 8103, antenna off and he had 121 feet- not too bad for conditions. With the engine running, he got only 38- one-third his regular range. Looking at his installation, he had a number of Giant Scale "no-no's"... his throttle servo was located in close proximity to his ignition system, plus his throttle pushrod was metal - both great ways to play havoc with your radio. Even though he'd had several successful flights, he wisely quit flying for the day. A re-routing of equipment brought his engine-on range to within 10% of engine off.
.
Poor range had nothing to do with range at all

A modeller getting ready to fly his nice new giant scale Bearcat came over to me and said he was having troubles. Seems he couldn't get but a few feet away from his model and control would become "crazy". Indeed, even though he had a PCM 649 receiver, he would get control interaction - his flaps would go down at odd times - none of the controls were "solid". I asked if we could get to his receiver, and he popped off a bottom hatch. When I plugged a freshly charged battery into his receiver, the system snapped to and worked perfectly. The culprit? He used another brand of switch harness whose lightweight leads deprived the receiver of its correct voltage. The moral: Stick to quality JR components. They cost more, but they work.

Download Your Own Log

Download your very own Range Test Log. You will need Adobe Acrobat to open this file

Download Range Test Log (11kb