---------------------------------------------------------- AeroElectric-List Digest Archive --- Total Messages Posted Sat 03/26/05: 10 ---------------------------------------------------------- Today's Message Index: ---------------------- 1. 05:58 AM - Re: Staco led dimmer (CardinalNSB@aol.com) 2. 07:22 AM - Re: Re: Diodes across relay coils (Paul Messinger) 3. 08:09 AM - Re: Re: Diodes across relay coils (Gary Casey) 4. 10:10 AM - Re: Diodes across relay coils (Eric M. Jones) 5. 10:39 AM - Re: Re: Diodes across relay coils (Joel Jacobs) 6. 10:54 AM - Re: Re: Diodes across relay coils (Joel Jacobs) 7. 01:23 PM - Hall effect current sensor (=Van's Ammeter for sale ) (sarg314) 8. 01:54 PM - (Eric M. Jones) 9. 04:04 PM - Re: (Joel Jacobs) 10. 07:25 PM - Re: Re: Diodes across relay coils (Paul Messinger) ________________________________ Message 1 _____________________________________ Time: 05:58:22 AM PST US From: CardinalNSB@aol.com Subject: AeroElectric-List: Re: Staco led dimmer --> AeroElectric-List message posted by: CardinalNSB@aol.com The EDMO flyer said that Staco is now making led replacement bulbs with no external resistors required for 6 or 12 or 24 volt, to fit their lighted switches. It also said they are making a led dimmer, but I couldn't find anything on their website and nobody had any information. Anyone know about this? skip simpson ________________________________ Message 2 _____________________________________ Time: 07:22:27 AM PST US From: "Paul Messinger" Subject: Re: AeroElectric-List: Re: Diodes across relay coils --> AeroElectric-List message posted by: "Paul Messinger" Technically very correct as one needs to decide what is important. Contactor contacts welding shut or a long opening time can be the decision driver. These side affects are real. 20V negative spikes are very unlikely to be a problem and never a problem in well designed electronics. In fact many (most nearly all ???) modern solid state devices have built in ESD and reverse voltage protection built in that works for input energy far far in excess of the subject 20 V negative spike. Modern electronics are properly protected to the small negative spike you are talking about. Also other than the contacts or solid state device doing the opening of the relay coil there is no reasonable way for anything to even "see" this spike. Its a trade off between things to protect. My point is the industry expert have all spoken and say back to back Zeners are far and away the best and the dimple rectifier is not very good at all. As I have often said there are several good solutions and many bad ones. I feel that the trade off in this case is simple. Contact welding or contact bounce that produced large current spikes are better stopped than a worry about a 20 V spike on the coil leads. What is really important is to know about this and make your own decisions. How many on this list knew that the relay industry felt so strongly about diodes were bad and back to back zeners were so good in protecting the quality of their product?? Paul ----- Original Message ----- From: "Ken" Subject: Re: AeroElectric-List: Re: Diodes across relay coils > --> AeroElectric-List message posted by: Ken > > But a diode clamps the negative pulse to about 1 volt. Seems to me that > the difference may be significant to some electronic circuitry that is > sensitive to negative spikes. > Ken > > Paul Messinger wrote: > >>--> AeroElectric-List message posted by: "Paul Messinger" >> >> >>The bidirectional transorbs clamp the negative spikes to less than 20 >>volts >>and also have no side affects of merit like contact bounce which is hard >>on >>contacts (not to mention making noise and other potentially nasty things). >> >>snip >> > > > ________________________________ Message 3 _____________________________________ Time: 08:09:28 AM PST US From: "Gary Casey" Subject: Re: AeroElectric-List: Re: Diodes across relay coils --> AeroElectric-List message posted by: "Gary Casey" In all this talk about diodes it looks like some of the fundamentals have not been presented (I think you'll find them in Bob's book). The normal case is for a relay to be connected to the battery through a switch - the switch in this case is on the "high side". When the switch is opened the current attempts to keep flowing because of the inductance of the coil. Because one side of the coil is connected to ground the voltage on the other side will by necessity (I haven't installed a diode) go to a large negative value. Something has to give, so in this case there will be a momentary arc drawn across the switch. So what is the effect of this? The high voltage transient exists ONLY in the wire between the switch and the relay. It doesn't exist in any wire that powers anything else. So the inductive spike is NOT in danger of doing any damage. The relay coil can certainly handle it and the switch was presumably designed to handle this type of event. The other outcome of this type of switching is that an radiated EMI event will be created that could interfere with electronic equipment. Usually this doesn't matter as the relays in question are generally for the starter or master and in both cases the operator probably doesn't care if something is affected. The description of a "low-side" switching arrangement is basically the same, except that the voltage will go to a large positive value. If one really wanted to protect the switch, keep the relay working correctly and eliminate EMI the best approach is probably Zener protection in parallel with a capacitor. I'm not sure any of this is really necessary unless there is a relay that is used often during flight (gear motor contactor?). Much ado about nothing? Gary Casey ________________________________ Message 4 _____________________________________ Time: 10:10:05 AM PST US From: "Eric M. Jones" Subject: AeroElectric-List: Re: Diodes across relay coils --> AeroElectric-List message posted by: "Eric M. Jones" --> AeroElectric-List message posted by: "Gary Casey" >In all this talk about diodes it looks like some of the fundamentals have not been presented (I think you'll find them in Bob's book). Gary, We run the risk of parroting what we have learned and what is in the textbooks. The point has been made that this has been revised radically by the people who know. Please review the links previously posted. >The normal case is for a relay to be connected to the battery through a switch - the switch in this case is on the "high side". When the switch is opened the current attempts to keep flowing because of the inductance of the coil. Well, because of the collapse of the magnetic field. But not to quibble.... >Because one side of the coil is connected to ground the voltage on the other side will by necessity (I haven't installed a diode) go to a large negative value. Let's put some numbers on this....typically 1000-1500 volts for a 12V coil. >Something has to give, so in this case there will be a momentary arc drawn across the switch. So what is the effect of this? The high voltage transient exists ONLY in the wire between the switch and the relay. It doesn't exist in any wire that powers anything else. So the inductive spike is NOT in danger of doing any damage. We part company here. The spike radiates into space and every wire, circuit board trace and conductor everywhere. They teach courses on the subject EMI, EMC, RFI, whatever. Besides the switch has one side attached to a positive bus that is connected to your pricey video game in your panel. >The relay coil can certainly handle it and the switch was presumably designed to handle this type of event. The other outcome of this type of switching is that an radiated EMI event will be created that could interfere with electronic equipment. Usually this doesn't matter as the relays in question are generally for the starter or master and in both cases the operator probably doesn't care if something is affected. There you go.... >The description of a "low-side" switching arrangement is basically the same, except that the voltage will go to a large positive value. The transient is always negative. > If one really wanted to protect the switch, keep the relay working correctly and eliminate EMI the best approach is probably Zener protection in parallel with a capacitor. Read the references. Potter and Brumfield says the best solution is a bi-directional zener-type device. >I'm not sure any of this is really necessary unless there is a relay that is used often during flight (gear motor contactor?). Hell, I didn't need the landing gear anyway.....think I'll practice my belly landing technique. >Much ado about nothing? Gary Casey You decide. That's why they call it experimental. As was mentioned before---nobody suggests tearing out all your diodes. But if you are building the electrical system you should carefully consider that there is an improved way of squashing transients. And if you have had a problem you might consider trying the new way to see if it solves your problem. The reason this was not done a generation ago was, that the technique was not known, the parts weren't readily available, and the requirements for "smooth power" were less demanding. Regards, Eric M. Jones www.PerihelionDesign.com 113 Brentwood Drive Southbridge MA 01550-2705 Phone (508) 764-2072 Email: emjones@charter.net In theory there is no difference between theory and practice. In practice there is. --Yogi Berra ________________________________ Message 5 _____________________________________ Time: 10:39:23 AM PST US From: "Joel Jacobs" Subject: Re: AeroElectric-List: Re: Diodes across relay coils --> AeroElectric-List message posted by: "Joel Jacobs" Couple comments, food for thought... Rate of change of current in an inductor = VT/L (Delta I) relay coils contain inductance AND resistance. Lets assume 12V 120 ohm coil and simple catch diode. When you first close the switch you have 12v accross the coil and zero current. The 12V accross the inductance causes the current to increase. As the current increases there becomes a voltage drop accross the coils resistance that subtracts from the voltage accross the inductance. Huh? Yes, When the current stabilizes it is because the whole 12v is dropped accross the resistance and there is 0 volts across the inductance. See - if there were still voltage accross the inductance the current would change right? In our case if the current stabilized at say 50ms we could say that we applied 12*.05/2 or .3 volt-seconds to the inductance to energize the coil. To de-energize the coil will also require .3 volt-seconds in the oppisate polarity. When we open the switch the current wants to keep going in the inductance and will keep going until .3 volt-seconds has been applied by the spike. With no catch diode the spike is limited by the interwinding capacitance of the coil. As we see the greater the amplitude of the spike the faster the inductance resets - if we allowed the spike to go to 1000v the coil would reset in .3/1000 = 300us - that might be a little hard on the switch though, but you can see that a higher voltage spike WILL reset the coil faster. With a simple catch diode it limits the spike to about 1 volt and you would think it should take .3/1 = 300ms to reset right? I don't think so. See at the moment you open the switch the diode is clamping the voltage accross the coil at 1 volt but there is still 100ma flowing through 120 ohms and that HAS to be 12v. Subtract the 1 volt clamp and it still leaves 11volts across the inductance to reset the coil. I would suggest that the energizing and de-energizing times with a simple catch diode may be nearly the same. Will using a transorb in place of the diode increase the spike and cause the coil to reset quicker? Absolutely without doubt! Will that decrease contact opening time? Maybe, maybe not. If the coil resets before the contacts open then the open time is determined by the spring and the inertia of the contact. I can't see how resetting the coil faster at this point could reduce opening time. If the coil is not fully reset when the armature begins to move then resetting the coil faster would definatly improve open time as the residual magnetisim would be fighting the spring. I would ASSUME ( yeah I know) that relay manufacturers would have designed thier coils so that they would reset before the armature starts moving using a standard catch diode. Joel ________________________________ Message 6 _____________________________________ Time: 10:54:09 AM PST US From: "Joel Jacobs" Subject: Re: AeroElectric-List: Re: Diodes across relay coils --> AeroElectric-List message posted by: "Joel Jacobs" ----- Original Message ----- From: "Eric M. Jones" Subject: AeroElectric-List: Re: Diodes across relay coils > >The description of a "low-side" switching arrangement is basically the > same, except that the voltage will go to a large positive value. > > The transient is always negative. > No, it isn't.. It is always oppisate the applied voltage that energized the coil. In a low side switch, the low side of the coil is negative while the switch is closed and the spike will be positive when the switch opens. Joel ________________________________ Message 7 _____________________________________ Time: 01:23:39 PM PST US From: sarg314 Subject: AeroElectric-List: Hall effect current sensor (=Van's Ammeter for sale ) --> AeroElectric-List message posted by: sarg314 I made a Hall effect battery current sensor using the Amploc Hall effect transducer and a DC_DC converter to generate +12 and -12v. Two relatively cheap parts plus a 1 ohm resistor gives you 1 mv out for each amp of battery current without breaking the battery cable to insert a shunt. It works fine when connected to my Fluke multimeter which has 10 megOhm input resistance, but when I hooked it up to a Van's ammeter (IE VAM40) the meter just lays there. Apparently the Van's meter has thousands of times less input impedance and the Amploc transducer just can't deliver enough oomf to drive it. I guess I'll have to use a digital LCD panel meter that has high input impedance. So..., I have a virtually new Van's meter that I can't use. $20 buys it and I'll pay the shipping. That's half what you'd pay to buy it from Vans. Anybody interested? -- Tom Sargent, RV-6A, engine ________________________________ Message 8 _____________________________________ Time: 01:54:59 PM PST US From: "Eric M. Jones" Subject: AeroElectric-List: --> AeroElectric-List message posted by: "Eric M. Jones" --> AeroElectric-List message posted by: "Joel Jacobs" ----- Original Message ----- From: "Eric M. Jones" Subject: Re: Diodes across relay coils >> >The description of a "low-side" switching arrangement is basically the >> same, except that the voltage will go to a large positive value. >> >> The transient is always negative. >> >No, it isn't.. It is always oppisate the applied voltage that energized the >coil. In a low side switch, the low side of the coil is negative while the >switch is closed and the spike will be positive when the switch opens. Joel, We are probably in agreement, I was referring to the coil terminals. The diode in always in the same direction because the spike is too--opposite the applied voltage. Eric ________________________________ Message 9 _____________________________________ Time: 04:04:50 PM PST US From: "Joel Jacobs" Subject: Re: AeroElectric-List: --> AeroElectric-List message posted by: "Joel Jacobs" ----- Original Message ----- From: "Eric M. Jones" Subject: AeroElectric-List: > >>No, it isn't.. It is always oppisate the applied voltage that energized > the >>coil. In a low side switch, the low side of the coil is negative while >>the >>switch is closed and the spike will be positive when the switch opens. > > Joel, > > We are probably in agreement, I was referring to the coil terminals. The > diode in always in the same direction because the spike is too--opposite > the > applied voltage. > > Eric > Hi Eric, You're right, relative to the other terminal it's the same polarity. High side vs. low side - the spike occurs on the opposite terminal hence the polarity reversal. I was considering the spike at the switch referenced to ground.. We are in agreement. Joel ________________________________ Message 10 ____________________________________ Time: 07:25:10 PM PST US From: "Paul Messinger" Subject: Re: AeroElectric-List: Re: Diodes across relay coils --> AeroElectric-List message posted by: "Paul Messinger" Your are incorrect in your assumptions and I suggest rereading the references I originally posted. For example I tested several contactors communally used in our acft. With no diode it takes approx 10 milliseconds to open from switch coil opening. Add the diode and it takes 50 ms. Both times are to the first opening. Contact bounce is reasonably short and only one or two bounces with no diode. With a diode some opening bounces are numerous and include a spectacular arc at the end. The energizing times are not affected with any of the devices across the coil. The simple diode makes a huge difference. (see the referenced reports.) I have lots of proof from lab testing and its real. The coil resetting (as you say) is many times longer than the mechanical release. 5+ times in my testing. Relay manufacturers (as in my referenced links) ASSUME you know what you are doing and NEVER use a simple diode with out knowing the potentially very bad side effects that go way beyond any lengthen of the release time. That some relays come with diodes built in is demand not smart engineering. One thing we can all agree on is SOME type of suppression is required, the question is what kind. In my OPINION :-) its OK in many cases to use the diode as the relay contacts are not stressed (small current vs rating) and are not switching an inductive load. Paul ----- Original Message ----- From: "Joel Jacobs" Subject: Re: AeroElectric-List: Re: Diodes across relay coils > --> AeroElectric-List message posted by: "Joel Jacobs" > > I would suggest that the energizing and > de-energizing times with a simple catch diode may be nearly the same. > > Will using a transorb in place of the diode increase the spike and cause > the > coil to reset quicker? Absolutely without doubt! Will that decrease > contact opening time? Maybe, maybe not. If the coil resets before the > contacts open then the open time is determined by the spring and the > inertia > of the contact. I can't see how resetting the coil faster at this point > could reduce opening time. If the coil is not fully reset when the > armature > begins to move then resetting the coil faster would definatly improve open > time as the residual magnetisim would be fighting the spring. > > I would ASSUME ( yeah I know) that relay manufacturers would have designed > thier coils so that they would reset before the armature starts moving > using > a standard catch diode. > > Joel > > >