outacontrol41

My Feedback: (9)

After messing around with my McDaniels glow driver and growing very frustrated from it's inability to keep the glow plugs warm or even start my O.S. FR-300 Syrius radial engine, I've decided to go back to the user's manual and construct a basic system with adequate amperage and voltage to keep all plugs lit during flight. I knew that these types of multi cylinder engines need much amperage to keep plugs lit but I underestimated how much. The owner's manual says at least 10 amps and I was using nowhere near that with a 4200 mAh Ni-Mh battery and McDaniels driver. Here is my plan in the illustration... the batteries will activate the plugs from idle up to 1/4 or 1/2 of full throttle after which the switch goes off and cuts off battery power. The glow will not activate again until throttle is dowm to idle again. Is there anything that I missed or do I have a winner? I know this has been covered in other forums but here we go. Any advice will be appreciated.

Radial power

Hi outacontrol41,

in general your circuit should work.

HOWEVER try to do WHATEVER you can to minimize the circuit resistance. That's the absolute key point.

Given the amount of current that must flow and the low voltage, even a very small resistance can indice a big voltage drop and cause problems.
So, for example, try and use the heaviest gauge wire you can, especially in those sections of wire that must carry the combined current of all plugs. And try to keep the wiring as short as possible. As an example, I used 6mm^2 wire for the common wires for my FR5-300 (I have one too ).
Also, make sure that the glow plug connections and the crankcase connections are as good as possible electrically, otherwise they'll increase the circuit resistance and reduce the current.
And also I'd replace that switch. Mechanical switches often have a non-negligible resistance. It would be better to use a different type of switch, such as an ultra low resistance power MOSFET (Metal Oxide Semiconductor Field Effect Transistor). This MOSFET can then be driven by the mechanical switch if you want, or maybe from the output from an electronic RC switch.
And you also need to ensure that your battery can provide sufficient current without having its voltage sag too much. Even if you can make a super low resistance setup, but your battery voltage sags when under load, you're still going to have problems.

To test the performance of your setup, the logical thing to do would be to use a multimeter to measure the current flowing through the circuit. However this is not completely practical: most multimeters are limited to about 10A maximum current handling capacity, and in any case you'd need to insert the multimeter in series with the circuit to take this measurement, adding the multimeter's "burden voltage" to the circuit and thus reducing the current.
Instead you can take voltage measurement across each component of the system: start by measuring the battery voltage under load, and then measure the voltage drop across the wire going from the battery positive to the first component. Then measure the voltage drop across that component, then the voltage drop across the next section of wire and so on. In the end the sum of all voltage drops should be equal to the battery voltage. By doing this you can find out if there are "bottlenecks" in your setup and, most importantly, where they are.

Let's go to a practical example.For my test stand I have the following setup: the main glow plug battery is a 2V 25Ah lead acid battery. It's more than capable of providing the, roughly, 3A per plug that OS F plugs require. However its voltage is slightly too high to be directly connected to the plugs. So I use 5 10W resistors to slightly drop the voltage to the desired level. Notice that I used one such resistor per each glow plug lead, as suggested in the engine manual. The value of these resistors is determined using Ohm's law V=RI. With a 2V battery and a desired 1.5V across each plug, we must drop 2 - 1.5 = 0.5V across each resistor. Given that the normal current for an OS F plug is about 2.5 to 3A, we can rearrange the formula to get R=V/I = 0.5/3 = 0.16Ohm. The power dissipated in each resistor is given by P=R*I^2 = 0.16*3^2 = 0.16*9 = 1.5W. So I used 10W resistors to have A LITTLE BIT of safety margin Not a bad idea, actually, as they get quite hot.
Actually 0.16Ohm is not a standard value, so I increased it to 0.22Ohm to be conservative. In hindsight I should have dropped it to the next lower value (perhaps 0.15Ohm), as the rest of the circuit invariably introduces some resistance. I may change this in the future as currently the resistors drop more than 0.6V instead of the desired 0.5V.

All of the common wiring (i.e.: wiring handling the current of all plugs) is made with heavy 6mm^2 wire, while individual glow plug leads use 2.5mm^2 (if I remember correctly). This is needed because the 2V battery is located on the ground and quite far from the engine, so I have to compensate for the long wires using bigger wires. In an onboard installation this should not be a problem, but using larger wires is never bad (except for weight reasons...).
Initially I used the plug leads supplied with the engine to connect to the glow plugs, but these provided an unreliable connection. So I recently replaced the OS connectors with these http://www3.towerhobbies.com/cgi-bin...&I=LXBC36&P=ML . I can tell you that they are a pain in the butt to connect and disconnect, as the FR5-300 has its glow plugs installed very deep, so it's difficult to get the connectors in place or remove them. Also they come with a length of very thin wire, so that's not the best. Probably if I had to do this again I'd use these http://www3.towerhobbies.com/cgi-bin...&I=LXBC26&P=SM even though they are EXPENSIVE!

Lastly, the switch. I started out with a standard 16A mechanical rocker switch, but some multimeter measurements revealed that I was dropping several tenths of a volt through the switch. Only about 0.6V reached the plugs. Not good.
So I ordered some IRLB3813PBF power MOSFETs (there are many other suitable n-channel power MOSFETs one can use), and then built a small circuit (see the circuit diagram attached).
Basically there is a row of 5 MOSFETs (Q3 to Q7), each one with its drain connected to one glow plug (PAD1 to PAD5, each going through one of the power resistors, not shown in the diagram) and its source connected to the battery negative GND. The 2V battery positive is permanently connected to a lug on the engine crankcase (again not shown in the diagram).
The gates of all of these MOSFETs are driven by another battery, a 12V lead acid battery, that I use to drive the MOSFETs and a few LEDs, as well as to power the electric starter (if I need it). This 12V power goes through a green LED (to indicate 12V power) and its current limiting resistor R5 and also goes to an amber LED (and the current-limiting R4) that is grounded through a 2N2222 transistor (Q2). The base of that transistor is driven by the 2V of the glow battery (through a current limiting resistor R2). The idea is that the amber LED is powered by the 12V battery, but current can only flow is the transistor is turned on by the presence of the 2V battery. So the amber led indicates whether both 12V and 2V circuits are connected at the same time. If only one of the two is connected, then this LED stays off.
Lastly I used a small SPDT switch (S1) to control the circuit: one side of the switch connected to battery negative (by the way both batteries have their negatives connected together), the other side to 12V and the center is the output. This output goes both to the MOSFET gates and also goes through a blue LED, its current limiting resistor R3 and another 2N2222 transistor, again driven by the 2V battery through resistor R1. Again, the idea is that the blue LED is powered by the same signal that turns on the power MOSFETs, but the LED only turns on if the transistor is turned on by the 2V battery. So the blue LED indicates whether 12V is connected AND 2V is connected AND the MOSFET gates are turned on (that is if glow heating is on).



Notice that this circuit is completely unprotected: if one hooks up the 12V battery reversed, then BOOM! That would be easy to fix, though. A single diode in the 12V positive line should be enough. The 2V battery should be low enough in voltage that nothing gets damaged if it is accidentally connected backwards. And in any case that battery features two screw posts of different sizes, so it's fairly evident if the wrong connection is made.
Also, when compared to your circuit, mine is a low-side switch, while yours is a high-side switch. That is my circuit places the switch in the return connection for the current, the one that goes to the battery negative. Yours, on the other hand, places the switch along the connection going to the positive battery terminal. In your circuit you could move the switch where you want with no problem (that is you could build it as a high-side or low-side switch), but mine is desigend to be a low-side switch onyl. To make a high-side MOSFET switch, you'd need to use P-channel MOSFETS and rearrange a few connections. Doable but not worth it IMHO.

This circuit is most likely overkill for your needs, but you can use the idea of having a mechanical switch drive a much lower resistance MOSFET. Or a bank of MOSFETs, as in this case. The circuit can also be scaled up or down to whatever number of glow plugs is needed by simply adding or removing power MOSFETs. One could even use one such circuit to drive the plugs for more than one engine.

As for a flying glow driver design, I use a single power MOSFET (an IRF3711PBF, now being discontinued but I built that circuit several years ago) to drive both plugs of my Saito FA-90TS connected in parallel. Power comes from a single 10Ah 1.2V D-size NiMH cell. Given the much shorter wire run, I used 2.5mm^2 wire throughout. In this case the MOSFET is driven by the output of an electronic rc switch that is powered by the 4.8V NiMH receiver battery. There's very little drain on this battery, as it only needs to power the switch and a couple of LEDs (I like LEDs ). One thing to note in this case: the receiver battery voltage must be enough to fully turn on the MOSFET, otherwise the circuit will not work (and the MOSFET may burn out). The critical parameter is the MOSFETs Gate Threshold Voltage: for the IRLB3813PBF this value lies between 1.35 and 2.35V, so any voltage higher than this that is applied to the gate will turn on the MOSFET. A 4.8V battery is more than enough. even though the higher the better, as higher gate voltages slightly reduce the MOSFETs resitance. Other MOSFETs may have different gate threshold voltages: the ones having 4V or more may not work in this application, unless you can somehow provide a higher voltage to drive them.

You can find these techincal data in the datasheet of any MOSFET you may decide to use. For example here's the one for the IRLB3813PBF: http://www.infineon.com/dgdl/irlb381...5356603cfe258b

Hope this helps.