Replacing SMD ferrite bead on motherboard, what value do I need?

[rf473]

Hello,

I am trying to repair a motherboard that has no power on the PS/2 keyboard port. The problem is caused by a damaged ferrite bead. It does not have continuity across its terminals and when I temporarily bridge it with a wire, the keyboard is working again. The circuit leads to the 5V pin of the keyboard port, apparently, its purpose is to filter electromagnetic interference from the power signal.

I want to replace this ferrite bead, something I have never done before, and I have no idea about the correct parameters I should choose. Here is an example for an SMD ferrite bead:

Inductance: 0.047 µH
Q: 20
Self-resonant frequency: 320 MHz
Resistance: 0.15R
Rated current: 300 mA

Edit: As someone below pointed out this is wrong, these parameters are for an inductor, not a ferrite bead. A ferrite bead has parameters such as: Impedance at 100 MHz, max DC resistance, max current. Example: 200Ω, 0.25Ω, 500mA.

Could anyone please suggest which parameters I need? Attached is a photo of the motherboard with the ferrite bead marked in red.

[DavidAlfa]

Not really required, you can short it out and will work fine.

There was a old hack in where you removed the ferrite on the USB data lines and the speed greatly improved, I guess they filtered too much.

[rf473]

Yes, I have bridged it with a wire for now and it's working. But I still want to replace it, it is a vintage motherboard and for me it's important to retain its original condition.

[wraper]

Not really required, you can short it out and will work fine.

There was a old hack in where you removed the ferrite on the USB data lines and the speed greatly improved, I guess they filtered too much.

It will still work if filter on data lines is replaced with jumpers but improving the speed is an utter BS. There is no data rate reduction mechanism in USB, it either works at maximum speed or malfunctions and device disconnects.

[SeanB]

Easiest is to grab a ferrite bead from another dead board, or simply put in a zero ohm link in place of it, it will still look like a SMD part, but will slightly degrade EMC performance, which the case is doing a better job of degrading anyway.

[DavidAlfa]

It will still work if filter on data lines is replaced with jumpers but improving the speed is an utter BS. There is no data rate reduction mechanism in USB, it either works at maximum speed or malfunctions and device disconnects.

I know it sounds silly but it was true!
I can't remember the device, the speeds instantly went from ~20MB/s to 34.
Apparently it had too heavy chokes, so some packets had to be sent several times.

I'm not sure if that statement is true, "Works perfect or doesn't work at all".
https://www.totalphase.com/blog/2020/07/about-the-usb-protocol-common-usb-bus-errors-and-how-to-troubleshoot-them/

Bulk: General transfer scheme for large amounts of data.
This is for contexts where it is more important that the data is transmitted without errors than for the data to arrive in a timely manner. Bulk transfers have the lowest priority.
If the bus is busy with other transfers, this transaction may be delayed. The data is guaranteed to arrive without error.
If an error is detected in the CRCs, the data will be retransmitted.
Examples of this type of transfer are files from a mass storage device or the output from a scanner.

[wraper]

It will still work if filter on data lines is replaced with jumpers but improving the speed is an utter BS. There is no data rate reduction mechanism in USB, it either works at maximum speed or malfunctions and device disconnects.

I know it sounds silly but it was true!
I can't remember the device, the speeds instantly went from ~20MB/s to 34.
Apparently it had too heavy chokes, so some packets had to be sent several times.

It must be flawed testing method. Transfer depends a lot on a USB drive, file size and some other things. If there were enough CRC errors to cut speed almost in half, USB drive would lose connection all of the time.

[rf473]

I have read in some manuals about decoupling circuits that I found online. I did not understand 99% of it, but their examples mostly used values in the range of 200-600Ω at 100 MHz. Should I just buy some 400Ω ones and hope for the best? If that doesn't work, I will probably just replace it with a 0Ω part as suggested by SeanB.

[Traceless]

Measure FB17 and the one next to it. If they have the same/similar value use that as a reference for your replacement. Side-note: C304 right below FB17 seems to be in pretty rough shape you probably want to replace that one too.

[rf473]

Measure FB17 and the one next to it. If they have the same/similar value use that as a reference for your replacement. Side-note: C304 right below FB17 seems to be in pretty rough shape you probably want to replace that one too.

Do you mean measuring its resistance? This is the only thing that I can think of. Do I have to desolder it first? I have tested it and the resistance is about 0.25Ω. That would correspond fairly well to a ferrite bead range of 200-600Ω (smaller ones have a lower resistance, bigger ones have higher resistance).

[wraper]

Most of the time engineers place such ferrite beads just in case without ever finding the best value or figuring out if they are really needed. So there is no much reason finding an exact part.

That would correspond fairly well to a ferrite bead range of 200-600Ω (smaller ones have a lower resistance, bigger ones have higher resistance).

Ferrite beads come with different current ratings and DC resistance within the same package size and impedance rating.

[rf473]

No, it does not need to be an exact replacement. If it resembles what it originally did in that circuit, it's fine.

Ferrite beads come with different current ratings and DC resistance within the same package size and impedance rating.

I found that the ones with a rating of 500mA correspond fairly well to these values. PS/2 has a maximum current of 275mA, so I figured it could be one with such a rating.

[Traceless]

Measure FB17 and the one next to it. If they have the same/similar value use that as a reference for your replacement. Side-note: C304 right below FB17 seems to be in pretty rough shape you probably want to replace that one too.

Do you mean measuring its resistance? This is the only thing that I can think of. Do I have to desolder it first? I have tested it and the resistance is about 0.25Ω. That would correspond fairly well to a ferrite bead range of 200-600Ω (smaller ones have a lower resistance, bigger ones have higher resistance).

Since it is a ferrite bead you should measure the inductance. For that you will need at least one of those relative inexpensive component testers (link to Dave's review of one below, there is also a thread about those here in the forum), make sure to get one with SMD pads like the one in the video. Alternatively an LCR meter would do but is much more expensive. To measure the bead with such a tester it would of course need to be desoldered.

[rf473]

Since it is a ferrite bead you should measure the inductance. For that you will need at least one of those relative inexpensive component testers (link to Dave's review of one below, there is also a thread about those here in the forum), make sure to get one with SMD pads like the one in the video. Alternatively an LCR meter would do but is much more expensive. To measure the bead with such a tester it would of course need to be desoldered.

Thanks. It seems that I need to get me one of these, also for testing capacitors. I cannot desolder SMD components yet, but equipment is already on its way.

[wraper]

Since it is a ferrite bead you should measure the inductance. For that you will need at least one of those relative inexpensive component testers (link to Dave's review of one below, there is also a thread about those here in the forum), make sure to get one with SMD pads like the one in the video. Alternatively an LCR meter would do but is much more expensive. To measure the bead with such a tester it would of course need to be desoldered.

You cannot measure inductance of a ferrite bead and get a meaningful result which will help you to find a replacement. Their inductance is not rated for a good reason. Just use any ferrite bead with sufficiently low DC resistance to not drop too much voltage at max current, and everything will be fine. Anything beyond that is a waste of time and effort which won't give better results.

[wraper]

I want to replace this ferrite bead, something I have never done before, and I have no idea about the correct parameters I should choose. Here is an example for an SMD ferrite bead:

Inductance: 0.047 µH
Q: 20
Self-resonant frequency: 320 MHz
Resistance: 0.15R
Rated current: 300 mA

These are specs you get for normal inductors, not ferrite beads.

[Bud]

I found that the ones with a rating of 500mA correspond fairly well to these values. PS/2 has a maximum current of 275mA, so I figured it could be one with such a rating.

Use it and call it a day.

[tooki]

Most of the time engineers place such ferrite beads just in case without ever finding the best value or figuring out if they are really needed. So there is no much reason finding an exact part.

Tell me about it!
I spent half the day yesterday modifying a board for someone to remove and bridge all the SMD ferrite beads he added to his design “just in case” which then actually caused his board to be more susceptible to oscillation. (DC-4GHz RF amplifiers, four channels, with two stages per channel, in close physical proximity, despite being expressly warned by our RF expert that this is likely to oscillate… I didn’t design it, just did the assembly for him.)