Brian Parker

The Spektrum AR500 receiver.   Comparison with previous Screening tests (and slightly off thread).   The test was conducted with the Spektrum AR500 receiver in the screening box as per previous tests. The receiver passed  full range tests with the transmitter antenna at both horizontal and  vertical  on all but one occasion when it failed at 24 paces (with the transmitter antenna horizontal). The Long screened antenna was curled and in physical contact with the inside of the box, so the testing was repeated with the antenna padded and away from the sides of the box..it still (as anticipated) passed the range tests. Conclusion, either this is a very robust receiver or the screening box is useless as a screen.  I suspect the former.

I'll put an  AR500 receiver in the screening box and compare its 'range test'  to the others tested.

Small Stepper motor?

Chris,   My tests were all conducted outdoors and repeated several times (because results were not as expected).  The only one to vary (and then by only a small amount) was the Orange receiver (screened and without the remote).  The box is of Die-cast Alloy previously successfully screening a VHF module, not sure of its UHF/micro wave properties and at 6X3X2inches antenna installation just about OK, the antenna wires themselves were not individually supported. Testing was with the box 'earthed' to 0volts and also tried 'un-earthed'.  I initially expected to have a very weak signal or even no signal at all into the box.  It seems to me the system very robust and reinforces the view that the majority of failures are due to poor installation or pilot error rather than to system weakness.  Also the Orange receiver is a good buy.

I’m aware of what Spektrum state but my test indicated an increased gain with the remote connected and equally screened.  How else to explain the difference in range observed when only the main receiver was connected and screened.

Tim, Yes, I was surprised at the performance of the Orange Receiver. A test with an Orange remote (when they come down in price) as opposed to the Spektrum Remote on both main receivers might produce further unexpected results. More test results This time with the main receiver only in the screening box.  The Orange range test was almost the same as its previous test failing at 8-9 paces but the AR6200’s range was reduced to 9 paces.  The remote RF/antenna on the AR6200 must be increasing the gain of the main receiver/antenna ie. they must be working together (matched?) whilst the Orange receiver is looking at the each antenna separately and using the strongest (switching or locking? more testing needed). Results of the tests were not fully as expected but confirms that the remote receiver is RF active and will give added security to both main receivers. Also the Spectrum receiver combination performed better overall than the Orange receiver in combination with the Spektrum remote.  Performance under screened conditions was much better than anticipated throughout the testing.

     Some unexpected results. Equipment..        Spektrum AR6200 Receiver.                                 Spektrum Remote Receiver.                                 Orange 6 channel Receiver.                                 Multiplex Tiny-s Servo.                                 4 cell NiMH Battery.                                 Sealed High EMC screening Box                                 Spektrum DX6i Transmitter. Results.. a)      With Orange receiver + remote + battery all in box with servo outside box with only 2mm cable exposed. (Antenna closer to each other than recommended.) Transmitter at ‘Range Check’ Range test failed at 9 paces. b)      With Remote receiver outside the box. Range test failed at 16 paces. c)       With AR6200 +remote + battery in box and servo outside Range test failed at 14 paces. d)      With remote receiver outside box. Range test failed at 16 paces. Range testing (normal). Both receivers (18inches above ground) gave the Orange receiver a slight edge over the AR6200. Range testing both receivers with the remote receiver connected gave the AR6200 slight edge. A vertical antenna gave slightly better results than when horizontal. (But remember this was with the receivers close to the ground). Range test failed (as Expected) when the antenna was pointed directly at the receiver.   Tim, Regarding Park Fly receivers. It’s a bit like Performance Cars. You have to do a lot(and spend more) to gain that little extra. With receivers the problem is ‘noise’, selectivity and overloading of  the front end. From what I’ve heard Park Fly2.4Ghz receivers seem to have more than ‘Park Fly’ range.   Edited By Brian Parker on 05/12/2010 12:09:34

   The signal is in all probability penetrating the screening.  To be fully effective the screen should be grounded. ie connected to the negative 0volt line.  The power feed cables are always a possible source of ‘noise’ at lower frequencies, but for a 2.4Ghz signal to reach the RF stages through the power cables and at the same time maintaining full integrity is extremely doubtful.  Intrigued by the results with the remote Spektrum receiver.  May try some tests to compare results. If OK with you Tim. Note. The remote will not have quite range of the main receiver.

  Tim, Did you re-establish contact with the main receiver before testing the remote receiver in your final test?

  Agreed, elevator trim it is. My CoG is tad over 4”(105mm) with an AUW of 2lbs.

  It’s all down to the 'Potential Difference' (PD).  Even with one battery discharged down to it’s safe minimum level of charge and with the others part discharged the PD between them will not facilitate a high discharge rate.

  I fail to see a problem provided all batteries are the same capacity and cell count. As I see it.  The charger  will initially start to charge the most discharged battery until it matches the next most discharged battery,  then continue to charge those two until they match the next most discharged  battery etc…etc…then continue to charge until all are fully charged.  Connected up but without the charger charging  the batteries would balance each other out.

  Contact Anglian Model Centre for a 60inch  Swordfish kit.   Traditional  kit (CAP).  £185.

  A couple of points.. Annealing of carbon steel is done out of contact with air, eg.  In pre-heated dry sand.   Cooled slowly in air carbon steel is normalised (de-stressed) not annealed.    As a guide when heating carbon steel … the point of carbon alloy transformation is when the steel no longer attracts a magnet. The required hardening temperature is just  higher. This temperature should be held for a few minutes.    To temper after hardening, polish with wire-wool then heat to a straw colour and immediately plunge in cold water and agitate. As a guide, this is approximately the melting temperature of lead-free solder.  

  This is probably caused by the level of decoupling provided within the PCBs of the E-Power  servos allowing unwanted common mode currents to conflict with the wanted signal (shared through the Y- lead).  When ‘paired’ with a servo from another manufacturer this servo provides the necessary decoupling.  If this is so, then a ferrite bead threaded on each Red(5volt) cable close to the servos together with  capacitors (1 to 10nf ceramic) connected across the 5volt and 0volt cables(again close to the servos  and with short leads) will provide decoupling through the  series low impedance now provided.  On the other hand it may simply be a poor joint at the y-leads common 0volt connection.  

Steve,  The cores will be OK just stacked.    My view for what it’s worth.. Interference from the switching regulator will appear on both the signal wire (white or orange) and the ground (black or brown) in phase. Hence 'common mode'.  The wanted signal is differential with respect to ground. The ferrite choke acts as a potential divider and with the stray circuit capacitance forming the shunt, will drop the unwanted (interfering) voltages across its large series impedance. Wanted (differential) signals will pass.  ie. The choke only stops the unwanted signals (that have been picked up on the leads) from entering the circuit.   An Alternative method.. a ferrite bead on the signal (white or orange) wire and a 10nF ceramic capacitor (short leads) between the signal wire and ground (black or brown) will be  effective although you may need more than one bead. I’ve not had any contact with RC PCM so am not aware of the actual PCM encoding used but with PCM signals DC stability is critical.Edited By Brian Parker on 12/07/2010 19:04:12

Steve, How deep are the ferrite chokes, GC only list the inner and outer diameter. The critical factor is the depth.  Inductance is proportional to the length of the turns through the ring (ie the depth of the ring) and also the square of the number of turns.  I suggest (respectfully) that you need to increase the number of turns and perhaps also use larger rings.  The aim is to increase the series impedance and thusblock the unwanted common mode currents.  Basically the more turns the better, you can also stack the rings to increase depth.

Just for interest.  An easy and cheap way to reduce voltages ( at reasonably high current levels) is to use bridge rectifiers.   15 and 20 amp bridge rectifier (15x15mm and 20x20mm) about £1 each.     Connect in series to obtain the voltage required. ie. 5 in series will drop a 12volt (13.2volt) lead acid battery down to 6 volts. Bolt to a heatsink if using at max. rated current.  

You just need these.   Four 120mAh (1/4AAA) NiMh cells from 'Component-shop' at 75P each. Just solder up to make a 4.8volt battery. Make up a couple of spares and fly all day.

Peter,  As with many imported Chinese items these CDI units probably share a generic PCB across the various ‘brands’. So if you are correct then faulty units could crop up anywhere.   However in my experience invariably when two units fail one after the other the true fault remains undetected.   Although come the Chinese industrial revolution who quality checks the quality checks?

Gary, The noise from the unit shows that the unit is charging itself. If you have access to a multi-meter you can test the sensor. With power on, the sensor centre wire(usually red) should show the line voltage (usually battery volts or perhaps slightly lower). Avoid shorting out the leads. Use a pin or similar down the sensor plug case to enable access to the centre cable with the (red) multi-meter lead, multi-meter negative lead to battery negative. If voltage is present, then transfer the pin/meter lead to the sensor signal wire (usually white or orange) and pass the magnet across the sensor. The meter volts will show a voltage rise and fall as the magnet passes if the sensor is OK.  If line voltage is absent then the cable or the unit is faulty.  If line voltage is present then the fault is with the sensor. (I take it you have tested the magnet?). 'Just Engines' supply replacement sensors for about £4.  I also see from their Web site that they  stock a ‘CDI’ tester for £7.99 Hope this helps, good luck. You seem to have spread your postings across two threads, perhaps a moderator will consolidate them into one thread.

You don’t give any information on any diagnostic testing carried out. Are you sure it’s the CDI unit that’s at fault. The sensor is probably the weakest link and is easy/cheap to replace if you don’t have facilities for testing it. Have you tried a replacement spark plug? A plug can breakdown under load conditions but still give a good spark under static field test conditions.  I take it that carburettor fault/fuel starvation has been eliminated. To have two new CDI units fail is probably quite rare.

Interesting, but..  Switching the transmitter off then back on will initiate a scan/reset in the transmitter and a GUID scan in the receiver ie. seconds worth of time and considerably longer than just a signal loss and recovery by the receiver.  In my opinion (for what it’s worth), simultaneous complete loss of signal in both channels in the receiver is very unlikely with correctly orientated dual diversity antenna (as you have demonstrated with your ground testing Peter.)  Plus a model in the air will have higher operating efficiency antenna than when on the ground.   The ‘Curly’ antenna configuration is probably quite common and possibly OK provided the general orientation rules are followed.  But it may also be the root cause of recorded failures.

    He trades as 'nilesinstall' and has several useful items.Edited By Brian Parker on 18/06/2010 11:14:30

Still off the actual thread topic (sorry) but.. A useful temperature module for your ESC, battery etc. (Handy also for IC temp measurement). From ebay and much cheaper than you can make yourself. It is also able to measure voltage, is programmable and has audible warning.Edited By Brian Parker on 18/06/2010 10:28:09