Originally Posted by awrightbrother

Do you mean, block off the EDF fan exit or the PC cooling fan exit?

Block off part of your EDF exhaust. this way you'll be forcing the motor and fan motor to work harder. This will put you into the upper levels of current draw and power usage. By doing this, you force The ESC, Cap Pac and Battery to work at their hardest. The harder they work, the more they dissipate power internally and the hotter they get.

Originally Posted by Peacemakr40

I'll put something together for ya so you can see the arrangement and spacing.

Thank you so much. I appreciate the insights you EE's can give. There are so many little eccentric issues for EDF's and model airplanes. It's nice to have someone who is into this stuff from an end user perspective and also can "very simply" (for people like me) explain best practices for EDF's in particular using the batteries and components we work with, and in the environments we work in.

So is there any point to adding caps if i can't get them any closer than 8" from the ESC?
I put my escs behind the fan as in pics.

Originally Posted by Doug Bateman

So is there any point to adding caps if i can't get them any closer than 8" from the ESC?
I put my escs behind the fan as in pics.

There's no point in adding caps other than to protect the batteries if you can't do a slow start\medium spool wind up of the rotor via the esc because the slow start would address a good portion of the "burst" issues that causes ripple voltages.

If your ESC components are performing correctly then components with 28 volts with 10% over voltage should not be doing 10% under those voltage and starting fires.

If you slow wind the rotor via esc or radio or thumb discipline (yeah right, not for me) then transient power is addressed and the pull from the packs is a lot smoother and adds a bit of scale to the sound to boot.

If you're esc can't do a slow start\slow spool of the motor then even at 6 inches putting caps on the wire leads will allow the "burst" power to be pulled from the caps and allow the battery to run a bit cooler and longer in flight because the "burst" power wont cause voltage dips and cut offs as fast as it woulld if you were running them with no caps.....just a little longer...not too much

Multiple smaller capacitors have better transient response than many other sources have but they can't store the energy as long as the other energy sources can...capacitors discharge fairly quickly even when not under load

maybe we ought to just put a ton of caps on so we can reallly get some power and longer flights

Just don't touch the wires with your tongue, those caps are sitting there ready to bite. I did that once, but I can't remember why.

Its all good info here and I've been wondering if the configuration I have (4x3s) is going to be of concern given all the batteries have different lead lengths simply due to space issues. One of the bats is up in the nose and its probably more than 12inches on that segment.

But again I think there would be value in the data acquisition software giving some indication about the ripple observed being "out of spec". This would be a huge value add for having the Castle data logging capability. And no, I am not thinking Castle is out to get us. If they really wanted to get us they'd enter the SHV market.

I'll ask again to the electrical engineers this time, could a person describe voltage ripple as water hammer in laymens terms?

Originally Posted by ApexAero

I'll ask again to the electrical engineers this time, could a person describe voltage ripple as water hammer in laymens terms?

No and if we use water as an example then:

1. Voltage is just like water pressure and like water pressure it is the force that is doing the pushing.

2. Current is the amount of flow and like water once it is flowing it doesn't want to stop.

3. Resistance is like a valve or a reducer.

4. A Capacitor is like the risers or water towers that you would use to get rid of water hammer. The caps absorb current until the cap is charged. If the voltage drops or changes direction the caps release the energy back into the circuit.

5. An inductor works similar to a cap but absorbs voltage until the magnetic field that the current creates is at max. And just like the cap, if the current drops, stops, or changes direction the current stored is released back into the circuit.

So voltage ripple is just what it sounds like, ripples, just like little ripples in water, and in this case the ripples are insignificate, even with pluse width modulation (PWM), because of the size of current being used. There can also be 2 types of spikes, a voltage spike and a current spike. Water hammer in electronics is more like a sudden disconnect. When you have current flowing in a circuit and suddenly disconnect the flow you get a spark, just like when you pull the plug a boiling kettle. And just like a very large water hammer, the water will burst the weakest point on the line. So when current is flowing and you disconnect or stop the flow, the current wants to keep flowing but can’t, so the electrons pile up and the voltage rises (voltage spike). In the case of an ESC, when you go from full throttle to off the current wants to keep flowing but can’t so it piles up against the FETs and because the current flowing in the conductors is so large we have a large magnetic field in the wires and when that gets put back into the circuit the voltage rise at the FETs can be so large that the FETs fail (Get Smoked), now you have your water hammer effect.

The caps will slowly absorb the current spike and slowly put it back into the circuit. No voltage spikes – no blown FETs. The longer the wire the larger the inductor. I forgot to mention that caps work the opposite of resistors when it comes to putting them in series and parallel. In parallel C1 + C2 = C Total, so 220uf + 220uf = 440uf. In series 1/C1+ 1/C2+1/C3 = C total, so 1/220uf + 1/220uf = 110uf. Also you can go to large of cap and cause problems slowing down the phase switching, which in turn can cause loss of power (torque) or even prevent the motor from turning but, you’d have to go really large cap/s for that.


Ray

Okay, so we should shoot for 30 to 50mV of P-P ripple. Is that what I am understanding after reading through the entire thread?

P.S. Found that Castle recommends ripple peaks should be less than 10% of the total pack voltage. Now that's 4.5v to 5v for most of us assuming 12s. Less is better of course but they go by Vpeak FYI

Question, I've seen the CC graphs posted from various guys and one thing seems the same, that the ripple looks to follow stick movements. After reading the info here I'm understanding that a cap addition would also help with this? So the info is saying the ripple is heavier with sudden stick movements, less with smooth movements from what I gather.

Originally Posted by db_sonic

Okay, so we should shoot for 30 to 50mV of P-P ripple. Is that what I am understanding after reading through the entire thread?

P.S. Found that Castle recommends ripple peaks should be less than 10% of the total pack voltage. Now that's 4.5v to 5v for most of us assuming 12s. Less is better of course but they go by Vpeak FYI

0.030 to 0.050 volts is very low and would be hard to achieve on the battery side. If you look at the data-logger files the ripple is measured by 0.2 volt increments and I believe that is on the motor side. On the motor side go with what Castle suggests. Remember, even though you are using a DC brushless motor, it is being run as a 3 phase pulsed DC motor so if you were to eliminate all ripple from the motor side the motor wouldn't run. The ESC is switching full battery power every time it pulses the motor so the motor leads are getting a lot of current. The ESC only changes the duration of the pulse and the timing of the pulse. More duration and timing gives you more power and speed. One last thing, on some senorless controllers the ripple can be used as feedback allowing the ESC to make adjustments.

Originally Posted by ApexAero

Question, I've seen the CC graphs posted from various guys and one thing seems the same, that the ripple looks to follow stick movements. After reading the info here I'm understanding that a cap addition would also help with this? So the info is saying the ripple is heavier with sudden stick movements, less with smooth movements from what I gather.

Yes and yes. The quicker the magnetic lines of flux cut across the wires the higher the voltage spikes (ripple).

Ray

db sonic:
So the info is saying the ripple is heavier with sudden stick movements,

Exactly...........as in.....don't blip your throttle....especially while on the ground as there is no air flow and the blipping causes heat in the controller in the form of ripple current. This told to me by CC tech.

So! I got a hold of a CC 160 HV and can't figure out for the life of me what type of caps are on the dang thing. I'm just wanting to go to Radio shack and get what I can find. I see the big ones are 470uF 50v, I can't find a type on them.
Ron101, do you still have a stash of caps?
Tom J

they need to be low esr caps I bought Panisonic 470uF 50v low esr caps like these
http://www.headfishop.com/panasonic-...itor_p988.html