Ferrite Snapon Question

Go To Last Post
22 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm supplying 5V to my circuit board via a 16-gauge wire . I'd like to put a ferrite core/choke snap-on on the wire but I'm only seeing two types of solutions out there:

1. The barrel style that seem to have a minimum of 5mm ID or so. That seems pretty big for my 16-gauge wire.

2. The ribbon style which I don't think would fit around my wire.

Anyone know of a product designed for what I have? Also, should I put the 16-gauge common wire through the same ferrite choke or in its own separate choke?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Laird Technologies specializes in the ferrite chokes and Digi-Key has several on this link under 5mm.

I have never used one, but it appears that both conductors travel through a common shield, but I will let the other Freaks check-in on that one!

Just some guy

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
1. The barrel style that seem to have a minimum of 5mm ID or so. That seems pretty big for my 16-gauge wire.

Ah yes, but it will allow many more turns. The inductance is proportional to N^2. Perhaps you need more inductance!/?

Does your fractional turn (which is effectively 1 turn) do the filtering that you are after.
What is the actual filtering that you are attempting?
What is the actual problem you are trying to fix?

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Ah yes, but it will allow many more turns. The inductance is proportional to N^2. Perhaps you need more inductance!/?

If I have a few turns through the toroid, the AVR wont start up.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That is still really puzzling.

Ferrite EMI suppressors are really lossy, and they should not ring. Something odd has to be going on.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

+1 There is something that does not make sense!

I am surprised that if it is a well designed board, that you even need any inductive noise suppression.
I am even more surprised, that increasing L stops the board from starting up.

Now to get to the bottom of it all.
I gather it is a home brew MPU board and the +5V supply is some distance away.
Which board do you have?
How far apart are theMPU & reg?
How much capacitive de-coupling at the regulator?
How much capacitive de-coupling where the +5V enters the board?
How much decoupling on the rest of the board?

Regulators have a lot if gain in them and can go unstable fairly easily if there is insufficient decoupling.
I am assuming that by increasing the inductance, you are changing the frequency at which it is taking off.

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It is a homebrew board. Not using a regulator, just 5V straight from this guy which is mounted 1 foot away: http://www.digikey.com/product-detail/en/VGS-25-5/102-1928-ND/2045659

No capacitor on the output of the power supply. There is plenty of decoupling caps scattered around the board per the AVR design notes.

I have added plenty of extra caps everywhere in the past to solve the glitch problem with no effect. I added a 47uF at the power in of the board. I added a humongous cap (in series with 10 ohms) at the output of the power supply. Caps never fixed the issue.

The problem is fixed if I add a choke at the 5V entrance to the circuit board. The problem is, if I put too many windings around the choke, the AVR wont start although 5V is present - the LED on the board that shows the presence of power glows just fine and my volt meter shows 5.1V.

As others have said, I think the AVR's power on circuit doesn't like the ramp up caused by too many turns on the choke and thus doesn't turn on.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A scope picture how the voltage rises with and without ferrite would still be nice. Most likely it rises too slowly anyway. Are you sure your AVR fuses are set to longest startup time possible?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yeah. I'll bring the scope with me next time to check it out. I usually use the longest startup time possible since I'm not sensitive to that. I can doublecheck that next time I'm there too.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Have you programed BOD to 4.3V ?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yup. I also see it work from time to time when a large contactor draws too much power. BOD works fine.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
The problem is fixed if I add a choke at the 5V entrance to the circuit board.

I asked this question earlier (in my first response) and ask it again now. What is the problem that you needed to fix in the first place?

Is this the first time you have used this board?
Ie. is it a proven reliable board?

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The problem I'm trying to solve is in my first post of this thread which is just me trying to find the best snap-on choke for a 16-gauge wire. That first post goes into all the detail but the gist is that I find chokes with either large inner diameters or chokes for ribbon but not for a 16-gauge wire (0.06" or so).

Later you mention more turns leads to more induction where I reply that I don't need more inductance as more turns prevents AVR turn-on.

You later suggest to add more caps to which I provide the experience of adding them in several places. I try to steer this thread back by saying the problem is fixed, I just need a choke physically designed for a 16-gauge wire.

Anyhow, I ended up getting 240-2146-ND and M8699-ND from Digikey as they had the smallest IDs. They aren't snapon but I can still install them easily. Thank you for your assistance. It is really appreciated.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I realize that your problem is trying to find a convenient core for your choke.
My question go deeper than that. Why do you even need an inductor in that power feed. If you don't need one, then your problem about getting a core vanishes.
You should not need one!

With the tests that you have done it would seem that you have a weird & mysterious instability problem.
One turn on a core might get rid of some RF radiation conduction either in or out of the board, and adding a few extra turns should not affect the MPU's operation whatsoever. Hence my suspicion that there is another gremlin!

So I am trying to did a little deeper as to what the underlying problem might be and understanding why you perceive the need for the choke in the first place helps.
Also knowing how much current the board draws might help in understanding why the board behaves the way it does.
Also what about the -ve ground return, where does that go? Shortest direct path? Twisted wires with the +ve wire? A picture might help!

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm fixing a problem where the AVR glitches when a 24V relay nearby opens. It doesn't happen all the time (perhaps 5% of the time) but it is definitely when the coil of that relay releases. The problem happens even when no load is going through that 24V relay so it's just the coil of that relay that is causing the problem.

There's a fast acting flyback diode on the relay coil. I put extra caps everywhere as well as a 5.3V zener diode at the 5V input to my AVR board.

The 5V wire feeds this AVR board and a couple of other boards that have basic stuff on them (shift registers, transistors, etc..) That means I have 5V and common wires going from here to there able to pick up any RF noise from that relay coil.

Nothing got rid of the glitch except a turn or two on a toroid I had lying around.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is there a possibility that your switching supply will not start properly with the added inductance? A voltmeter may well not tell you anything. Still, the added inductance is so small that it should not make a difference.

Some switching supplies will not run properly without a minimum load, although without the inductor it seems to do so. Put a 20 to 40 ohm resistor of sufficient wattage or a light bulb drawing a few hundred ma across the output and feed the 5 volts to the multiturn inductor going to the avr board. Any change?

Try it with a 6 volt battery with 2 diodes in series to bring it down to about 5 volts. What happens?

At this point you have not yet defined the problem. Until you do, you cannot truly fix the problem.

Rest assured that until you define and solve the underlying problem of WHY it is "not starting" with that inductor and supply, it is going to bite you in the ass when you least expect it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Rest assured that until you define and solve the underlying problem of WHY it is "not starting" with that inductor and supply, it is going to bite you in the ass when you least expect it.

Could not have said it better myself!
Even with some inductance it might glitch a lot less, by still "glitch" every now & then and your solution is only marginal at best.

What does the MPU do when it "glitches"?
Does it reset or just gobble up some incorrect logic level? If the latter, which input?

Can I assume that the MPU & 24 V relay have a common ground?
Do you decouple the relay coil with say a 10 Ohm resistor & say 0.01uF cap
Does the MPU operate the relay via a driver of some sort?
Does one 12 inch wire feed all the boards or are the
other boards wired via separate wires from the one power supply.

There will be a good reason why your circuitry is mis-behaving!

Picture?

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you all for the suggestions. I'll try some of those experiments as soon as I'm in front of the board again.

-The 24V relay and the MPU do share common ground.

-I don't decouple the relay coil with anything. It just has the flyback diode. The coil of this relay is controlled by a door switch to the cabinet and the pole of another relay. I can get the glitch to occur by tripping the door switch back and forth.

-The 5V wire goes from the power supply to the two boards - one that has the AVR on it and the other that has the shift registers and other stuff (no microcontrollers)

In general though, since the glitch only occurs when the relay opens (load or no load on that relay), can't I say that the magnetic field collapsing on that coil is causing the problem? Feels to me like I have isolated the problem and I just need to get the right choke.

Anyhow, I'll try the other experiments and report back soonest.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
can't I say that the magnetic field collapsing on that coil is causing the problem

You could, but unless you are 100% certain, I wouldn't.
For example ( without knowing the circuitry), a relay contact will produce contact bounce. If the signal goes direct to the MPU or even worse if it goes to the board with the shift registers and then to the MPU, you might be getting the effects of contact bounce which your software is not handling properly (perhaps even causing a reset especially if the code is written in ASM (a stack error is more likely in ASM than C).

Any amount of C's or L's will never fix the underlying problem in that case.

Certainly the MPU not starting with some extra L on the power feed needs to be fixed, as it is not correct

PS. I will respond to your PM in due course.

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Much of the mystery is where and how the current flows. Thus the physical layout of the interconnections is critical. I lost count at my previous job where the electrical engineers ignored the 'rules'.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This can be compared to not thinking of analog ground when you use a ADC.