The AM Forum
March 28, 2024, 08:59:23 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Protection from 27vdc coil relay trancients. Help is needed  (Read 12338 times)
0 Members and 1 Guest are viewing this topic.
AMLOVER
Member

Offline Offline

Posts: 114


« on: February 19, 2011, 07:34:32 PM »

What is please the right way to avoid the trancients from my 27vdc vacuum relays coils and from my antenna changeover relay coil?
A snubber or zeners back to back?
Where is the ballpark rates?
Thank you.
Stefano
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #1 on: February 19, 2011, 08:29:33 PM »

a diode in parallel with relay coil. Cathode to positive side anode to negative side. 100 volt PIV minimum.
Logged
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #2 on: February 19, 2011, 09:12:38 PM »

a diode in parallel with relay coil. Cathode to positive side anode to negative side. 100 volt PIV minimum.

  Although this is the convention among many, this approach has a drawback. The relay magnetic field will decay based upon the L/R time constant, and for big power relay that means a delayed release time, and probable contact arcing from the slow release. This may not be an issue if the sequenced timing of the T/R system is such that RF is removed before the contact starts opening. I like a big relay that "klunks" on both make and break. Adding a diode per convention often gives you a "Klunk" on make, and a masked "Thud" on break.

  As a test just hook up the relay to 27vdc, and let the wires make and break. Then connect the diode. There is a difference. So how to suppress the transient all depends on how you drive it, and the sensitivity of the driver to the transient.

  For example, driving the relay with a power fet as a low side switch does not need any further suppression as long as the fet is overrated for the task (maybe a 5A 200V TO-220 unit). Right after the relay is switched off, the relay coil collapses, and in doing so the +/- across the coil switches polarity. This momentarily raises the fet VDS until its rating is exceeded, and the transient is absorbed by the fet. This sounds brutal, but the fet makers actually allow this since fet's are immune to secondary breakdown (from a non-periodic momentary transient) unlike the bipolar transistor that can be ruined with a single VCB avalanche pulse.

  Another way is to use the diode as per convention, but in series with the diode insert a resistance where the resistor value is equal to the DC resistance of the coil. This will speed up the decay of the magnetic field at the expense of more transient voltage than just a diode alone.

  This issue was discussed last year in the QSO section concerning the runaway Prius thread. The automakers were accused of causing unintended consequences by omitting the conventional diode across each power relay, when instead the low side drivers were designed to absorb the transient. Here relay contact life was a motivating factor where a delayed contact release would accelerate relay contact pitting. Still, putting harnessing between the drivers, and the relays could be a problem since the high Dv/Dt transients might couple into other wiring.


Jim
WD5JKO
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #3 on: February 19, 2011, 09:20:19 PM »

A fair sized relay coil will generate a 1000 volt pulse when it opens. In time it will take out an active device not rated for that voltage.
Logged
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #4 on: February 19, 2011, 09:42:38 PM »

A fair sized relay coil will generate a 1000 volt pulse when it opens. In time it will take out an active device not rated for that voltage.

The first part is true for sure. As a kid I used to take tiny Radio Shack 9V relays connected as a oscillator such that it 'buzzed',  then I'd collect the back-emf voltage through a NE-2 and charge up a capacitor, like a 1 uf 400V. The NE-2 would extinguish once the capacitor was charged to several hundred volts. I built and sold these in the 9th grade to kids wanting to shock their friends.  Grin

The FET application I mentioned works because the fet will start leaking as a dissipative resistance when it's VDS rating is exceeded, and the 1000v transient will be clamped at whatever voltage the fet leaks at. This transient will dissipate very quickly and not harm the fet since the average power is almost zero averaged over several seconds. Don't try this with a bipolar transistor however, because the collector to base junction will short out from secondary breakdown (a failure mode that fets are most immune to). Like I did say though, use a beefy fet, not a tiny TO-92 job.

I ran into this issue on the job once where I needed to switch 28 VDC to a reeling system used to deploy a winged vehicle from an aircraft. That 28V relay was the size of a brick, and the single diode convention resulted in fried contacts that fused together leading to the reel system failing and overriding all interlocks. bad day.  Huh   The solution was to suppress the relay with a large 48v rated varistor. Here the driver was protected, and the large relay thumped quickly on both make and break.

Jim
WD5JKO
Logged
w1vtp
Member

Offline Offline

Posts: 2638



« Reply #5 on: February 19, 2011, 10:03:21 PM »

I have always put a diode across a dc coil on a relay to shunt off the back EMF that happens on removal of DC from the DC coil.

Al
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #6 on: February 19, 2011, 10:04:31 PM »

 True a Zener or transorb allows the voltage to fly up to the clamp voltage causing the relay to open quicker The higher the clamp voltage the lower the circuulating current.
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #7 on: February 20, 2011, 12:32:14 AM »

Stefano, I listened to the 1650-1800 KHz band last night here in Athens.  That's quite interesting - sounds like guys having 2-way QSOs, interspersed with music, then going on to a broadcast for a while.

One guy had FMing of his signal really bad, and some were overmodulating a bit, but most sounded pretty good.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
Steve - K4HX
Guest
« Reply #8 on: February 20, 2011, 12:37:08 AM »

Cool. Sounds like 75-meters back in the 1970's.
Logged
Tom WA3KLR
Contributing
Member
*
Offline Offline

Posts: 2120



« Reply #9 on: February 20, 2011, 08:30:34 AM »

The first I ran into this problem was 1976.  A SMALL 12 V relay release upset the digital electronics/microprocessor system I was working on.  I found that the relay coil was generating an oscillating burst up to +400 Volts.  Not only can you destroy the drive switch but upset surrounding circuits.  For our situation , putting a silicon rectifier diode across the coil was a good solution.  A series RC snubber may be a solution.

Jim WD5JKO is correct that the damper diode will prolong the relay drop-out due to the continued current flow through the coil by the diode.  Everything in engineering is always a trade-off.  What ever you do, put the suppression components at the coil, which is the energy source, to keep the current loop smaller.
Logged

73 de Tom WA3KLR  AMI # 77   Amplitude Modulation - a force Now and for the Future!
AMLOVER
Member

Offline Offline

Posts: 114


« Reply #10 on: February 20, 2011, 09:01:16 AM »

I usually energize the relays with a toggle switch manualy by setting them to ground as far the DC+ is always on the coil. The relays work perfectly but I am afraid that the trancients they create are dangerous for the sensitive output of my af mosfet modulator which is connected permanently on the mod transformer primary.
Is it possible to minimize somehow the huge trancient from the reactor when I shut down the transmitter other than the one I use now (see attachement) with the resistor that shorts the reactor few milisecs before the driver goes off?

W3JN, you mean that you heard us in Athens-Greece or you heard us in Athens-USA? the later will be interesting because even powerfull, our antennas are max 60-80ft high and this is considered nvis at 180m, with very high angle take off (70-90 degrees). Some go the low level mod and few like me the high level anode mod normaly with huge parallel and push pull triodes af amps like 4-8x833a or 2x3cx1500. I decided the solid state way but things are very dangerous with on-off trancients especially around the reactor...

Steve, yes it's true we go the 70s romantic way with some song in the begining and end of transmission for full frequency tests and for taking out the stress of the day. Sometimes with 5 to 6 stations on the circle it takes me 45 minutes till my turn comes, 3-5 times on the air is normal for 3 hours listening! Undecided
As you understand the most of our wasted power is for filaments and not for transmitting!

Frank, I'll try with the 1n5408 to the three of my relays and later with the zener, in case I don't find a good solution for reactors and relays trancients against my solid state af amp I'll move to a pushpull triode af amp 2x5868. In this case I'll need your driver to shake their grids, I hope you ship to Greece.
I had twice blown up the outputs of the solid state amp because of some trancients so I need to find a definite solution against that problem. Check please the schematic.

Thanks for any additional advice
Stefano


* trancient.JPG (125.97 KB, 1210x639 - viewed 532 times.)
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #11 on: February 20, 2011, 09:39:59 AM »

I live in Athens, Greece  Grin
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #12 on: February 20, 2011, 10:19:55 AM »

Stefano
et. al.

If you want the relay to open faster... and as an alternative to placing a resistor in series with the diode... you can place a Zener diode in series with the ordinary diode.

When the switch controlling the relay opens, the current in the relay's solenoid will flow through the Zener diode in series with the regular diode.

This will limit the voltage across the relay to the value of the Zener diode's breaksown voltage... and it will also place the Zener breakdown voltage in the loop through which the current flowing in the solenoid is flowing. This will dissipate the energy stored in the magnetic field of the solenoid.

di/dt = V(Zener)/L(solenoid) + i x R(solenoid)/L(solenoid) > V(Zener)/L(solenoid)


Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
AMLOVER
Member

Offline Offline

Posts: 114


« Reply #13 on: February 20, 2011, 10:52:44 AM »

Tom
Thank you, I'll do so and hope my problem will find its solution.

Jim
I'll study your replies and expertize about the right arrangement.
The thing is that I can't be sure where the trancient comes, I see only the result of its devil...
I'll supress everywhere and check again. Fortunately I have one more set of mosfets in case...

Al
Diode will be my first step, crossing fingers I hope this will save the amp's output.

Johny
You are welcome to join us at 1690khz anytime you feel boring, just turn your 160m transmitter to a bit lower and have ready some country music for entering in the circle  Grin

Thank you
Stefano
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #14 on: February 20, 2011, 10:57:40 AM »

Stefano - unfortunately I live in an apt in Filothea and xmitting, particularly down that low, is difficult.  I'll listen for you guys though!

edit.  whoever's on 1650 right now playing Greek folk music appears to be splattering +/-40 KHz, his audio goes away from time to time, but has a REALLY strong signal.  He sounds really good on channel, maybe the splatter is just because he's so strong here.  The guy on 1680 is sounding a bit rough, and I hear some carriers on 1690 but no audio as of right now.  The pop music on 1695 sounds pretty good.  The receiver is a Watkins-Johnson 9040 system with a WJ8626A HF module.

I didn't realize how many of these hobby stations there were!  I challenge some of you guys in the states with beverages aimed at Europe to copy these guys.  Might be tough with the extended AM BCB in the US, but maybe you'll get luck if condx are right, and they're on the air during mutual darkness in Eur/US.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #15 on: February 20, 2011, 12:14:03 PM »

A 1 amp diode is fine.
Logged
Steve - K4HX
Guest
« Reply #16 on: February 20, 2011, 12:15:52 PM »

Or you could put relays in the lines from the amp and disconnect the amp before turning off the final.
Logged
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10057



« Reply #17 on: February 20, 2011, 12:55:48 PM »

Cool. Sounds like 75-meters back in the 1970's.

Or 160m (1870-90 kc/s) last week.  I believe that station or those stations were on this side of the Atlantic, however.

Do the authorities in Europe make any attempt to shut down those "hobby" stations?

The nearly vacant 1700-1800 kc/s frequencies should be prime spectrum for hearing the signals in N America, since all but about one of the radiolocation beacons allocated to this segment have shut down, most likely with the advent of GPS. Apparently all the ones on 1900-2000 kc/s have closed down, except maybe for a few flea power floating buoy CW beacons.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #18 on: February 20, 2011, 01:50:26 PM »

I doubt you could get away with this in somewhere like Germany, but Greece is a different animal.  Greece has some laws that are unfamiliar to us.  The universities, for example, are autonomous regions that even the police nor fire dept can enter unless they are invited by the rector.  Broadcast stations in these regions are beyond the reach of the Greek equivalent of the FCC.

Don, I'm not hearing any of those guys in the 160 meter ham band.  There's a ton of them from 1600 up to about 1750 though, some with only 5 KHz separating them.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10057



« Reply #19 on: February 20, 2011, 03:05:15 PM »

I'm not hearing any of those guys in the 160 meter ham band.  There's a ton of them from 1600 up to about 1750 though, some with only 5 KHz separating them.

It should be fair game to hear them in N America on a good night, between 1700 and 1750, since there are few if any beacons on the air these days.  But the time differential may be the problem.  When it is dark on both sides of the pond, it is between midnight and 7 or 8 AM in Europe, and I doubt many of the "hobby" broadcasters would be on the air then, but maybe some of them would be willing to run tests for N American listeners.

I'm sure the one(s) on 160m the past few days were located here.  Some slopbucketeers accused me, since I came on the air with a QSO on AM 5 kc/s down, right after the music station cut out.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #20 on: February 21, 2011, 12:58:16 AM »

If you want the relay to open faster... and as an alternative to placing a resistor in series with the diode... you can place a Zener diode in series with the ordinary diode.

   This is correct, but i wonder what voltage zener is appropriate? Maybe equal to the relay voltage, or half that? Also when doing this we want to avalanche the zener, so the zener needs to point the opposite way as the diode (anodes connected together, diode cathode to relay +, zener cathode connected to relay -).

Jim
WD5JKO
Logged
AMLOVER
Member

Offline Offline

Posts: 114


« Reply #21 on: February 21, 2011, 05:08:59 PM »

Don,
we only transmit between 1620 and 1720 khz, majority of us are licensed hams that like some fun out of the legal bands so we push our oscillators a bit lower and meet the "traditional am pirates' that doesn't like broadcasting any more and prefer communication in this free in Europe "band".
I don't think you can get us as far we focus to the very local communication 300-500 miles max.
Johny,
In Holand & Belgium there are plenty of "pirates" in this "band" and many more in Fm band.
Italy is the haven of pirate radio.
What about a small verical coil antenna on the top of your building that you use for receiving and ....I say somenights have a little folklore fun with us.... Wink Pirating is a little salt in the ruled amateur life... 
Stu,
I am fast enough for the Greek level so not that important the relay fast switcing, the very important is to protect the af amp's output from trancients... Cry poor me! 
Steve,
For some reasons Stu noticed me not to shut down the Af amp and really there is not any problem with Af amp on...the problem is unpredictable and comes suddenly without knowing where from. On the end may be I'll start to shut it down even this will make me very slow because there is a 2secs step start in the af amp.
I corrected the tomorrow procedure, check the attachment.

Thank you all
Stefano


* trancient.JPG (139.53 KB, 1210x639 - viewed 532 times.)
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.065 seconds with 19 queries.