RCU Forums - View Single Post - What IBEC to buy?
View Single Post
Old 10-16-2011, 03:59 PM
  #58  
NJRCFLYER2
My Feedback: (42)
 
Join Date: Jul 2004
Location: Randolph, NJ
Posts: 878
Likes: 0
Received 0 Likes on 0 Posts
Default RE: What IBEC to buy?

Ifeel that it's necessary to dispellthe mistaken belief that a linear regulator can't be used in an IBEC application.It'smostly just a matter of putting a properly designed common mode filter network between it and the noise source.This has to beaugmented with additional low pass filtering, i.e. the supply current passes through a common mode transformer, whichis highly effective at cancelling RFI noise, along with the 2nd stage of the common mode filter, and thefilter capacitor bank in the low pass filter greatly assists the regulator in handling the current impulses when the CDI is firing. Thatassist tohandling current impulses isaccomplished with a mix of a fast, low ESR 100 uF part, and a pair of 470 uF electrolytics that handle the bulk of the initial impulse as the regulator responds. The regulator itselfalso hasto be selectedfor fast response to transients and has to have the PCB layout done properly to allow it to run stabley, and it should be a low drop out device as well.It's also good practice to place a one way opto-isolator in the signal path from the receiver to the IBEC, to assure that there is excellent Galvanic isolation in that path to the receiver electronics.

I've directly looked at the noisein the ground path, and have compared the Ultra IBEC with an olderdesign. That olderdesign had no regulation on it, and used a MOSFET transistor in the ground path to switch power on and off, which when turned on, allows whatever current flucuations andaccompanying noise thatwas let through the low pass filter to reach the receiver. It also seemed to rely somewhat on separatingit andthe CDI from everything with a very longextension from the receiver side, which isn't really the strategy I would pick, since a wire that longhas to get rolled up inside a small model, and is going to be somewhat unpredictable in it's effect on the receiver as a secondary antenna radiating RFI noise. The goal is to eat the noise a short distance from the CDI, not propogate it over a long length of copper.

In contrast, the Ultra IBEC output is very stable, recovers very quickly from each ignition impulse with very minimal ringing, and the noise that does reaches the receiver side is much lower that the other unit that I tested. Range tests prove this out as well.To be fair, this was an older design that I compared it against, and I have not yetlook at any new offerings to see if it has improved performance.

I've gota log of ground range tests performed during development,and have performedmuch inflight testingwithFM PPM and PCM receivers from the '80s to present day, andhave never had a probem with insufficient range with the models Itested. WhenI did inflight testing, itwaswith those same receivers, to the limits that youwould ever fly your modelin actual practice.I have had exactly one report of a range test on the ground thatdidn't pan out from a customer with an FM receiver, sohe didn't fly it. That Ultra IBEC was returnedunder warranty, and a failed solder joint was found to be the cause. There have beenfewer than 10 warranty claimsoverall. As far as spread spectrum goes, it's simply been bullet proof, with over 1500 units in the field in just under two years since its introduction.

I agree with one thing, which is that range tests should be performed, no matter what unit you buy. Depending on how an IBECtype device is designed to provide varying output voltage settings, it may be introducing additional noise of its own origin in the wiring path between the battery and the IBEC, in other words, directly through the receiver power bus. That is a possibility if a switch mode regulator/supply is used in the IBEC, because it canintroduce a high frequency chopping effect to the supply path as it performs its switching. Not saying it can't be done, but it's something to understand. If the rise times are very fast, it's possible to introduce higher frequency harmonics that have the potential of affecting things like the nearby frequency synthesisor inside every spread spectrum receiver. If the design handles that well, it might not beof anyconcern. So, do the range testing!