Also Id like to come back to the suggestion on adding small decoupling caps directly onto the LM3886 chips power pins. Im allready replacing the 100uF 50V supply decoupling caps with equivelent BG STD's. Due to their size and the grounding of the chip I cant really solder these directly to the chips pins, so they will just go in were the original caps are.
However, I can still solder in a couple of smaller caps direct on the pins as additional bypass. Im considering something like a small value Wimma ployprop or BG N, direct from the power pins to the ground pin. Whats the opinion on this idea? (Im leaning toward adding 10uF BG N's at the moment).
Mike.
However, I can still solder in a couple of smaller caps direct on the pins as additional bypass. Im considering something like a small value Wimma ployprop or BG N, direct from the power pins to the ground pin. Whats the opinion on this idea? (Im leaning toward adding 10uF BG N's at the moment).
Mike.
but the Auricaps did sound different. That surprises me.
But, do you want to take the risk without informing your insurance company?
Im not insured mate.
The thing is they are rated at 600v apparently (does a pair in paralell make the rating 1800v?), and I can solder them in after the fuses so If anything goes fubar the fuses will blow.
Anyone else have safety considerations?
Mike
The thing is they are rated at 600v apparently (does a pair in paralell make the rating 1800v?), and I can solder them in after the fuses so If anything goes fubar the fuses will blow.
Anyone else have safety considerations?
Mike
That 600V is probably a DC rating.
The mains gives ~360Vpk and spikes to 1 or 2kVpk are common.
That's why Xrated are tested to 1500V (a while since I read the X1 & X2 spec) and are designed to be self healing.
Parallel has the same voltage rating as a single.
A series pair if also fitted with voltage equalising resistors have about double the voltage rating of a single.
Is your home attached to a neighbour? Their insurance company may take you to court if you damage their home without insurance of your own.
The mains gives ~360Vpk and spikes to 1 or 2kVpk are common.
That's why Xrated are tested to 1500V (a while since I read the X1 & X2 spec) and are designed to be self healing.
Parallel has the same voltage rating as a single.
A series pair if also fitted with voltage equalising resistors have about double the voltage rating of a single.
Is your home attached to a neighbour? Their insurance company may take you to court if you damage their home without insurance of your own.
Ive got a bunch of decent varistors I can play with. How could I use these to make this mod safe?
mikesnowdon said:
I wonder if putting the cap in series with a resistor makes it more safe 😉
It would be helpful to know the true AC voltage rating of the cap though.
That said.... fuse your contraption and put it in a fire-proof box. That seems like self-insurance to me!
Simon
Hi Guys,
Stop messing with the mains power. It isn't time for that for one, and for two, it isn't smart to do without knowing the electrical code.
BTW, Andrew is right on the mains rated caps. They are another part I stock, 'cause they are different. They are also approved by our safety standards association.
Simon,
RC networks after the regulator do not always sound bad. This is for signal stuff. They are a wonderful way to reduce the amount of high frequency stuff on your DC supplies. RC networks before the regulator are very good for the same reason.
-Chris
Stop messing with the mains power. It isn't time for that for one, and for two, it isn't smart to do without knowing the electrical code.
BTW, Andrew is right on the mains rated caps. They are another part I stock, 'cause they are different. They are also approved by our safety standards association.
Simon,
RC networks after the regulator do not always sound bad. This is for signal stuff. They are a wonderful way to reduce the amount of high frequency stuff on your DC supplies. RC networks before the regulator are very good for the same reason.
-Chris
mikesnowdon said:
Anyone care to comment?
I know someone suggesteed adding 33uF Pana (FM's I think) earlier, but Im using Blackgates so any additional bypass cap needs to be equivelent or better quality. Or the right type of cap to make a difference.
Mike.
I would put a small polypropylene cap directly on the pins. A cheap poly cap is still better than a BG IMO
Cool,
What kind of value os small? something like a 1uf Wimma plyprop perhaps? Or plycarb maybe?
Mike.
What kind of value os small? something like a 1uf Wimma plyprop perhaps? Or plycarb maybe?
Mike.
Andrew.
I think you are referring to this, quoted from the LM3886 datasheet:
"Inadequate power supply bypassing will manifest
itself by a low frequency oscillation known as “motorboating”
or by high frequency instabilities. These instabilities can be
eliminated through multiple bypassing utilizing a large tantalum
or electrolytic capacitor (10 ìF or larger) which is used to
absorb low frequency variations and a small ceramic capacitor
(0.1 ìF) to prevent any high frequency feedback through
the power supply lines."
We are talking about the 0.1uF ceramic right?
also in the datasheet: (quote)
"It is advisable to keep the output compensation
components and the 0.1 ìF supply decoupling capacitors as
close as possible to the LM3886 to reduce the effects of PCB
trace resistance and inductance."
This indicates a 0.1uF soldered directly to he chips ppins would be good. I think this was suggested earlier in the thread?
Doses it have to be a ceramic or are there other types of cap which would be better?
Mike.
I think you are referring to this, quoted from the LM3886 datasheet:
"Inadequate power supply bypassing will manifest
itself by a low frequency oscillation known as “motorboating”
or by high frequency instabilities. These instabilities can be
eliminated through multiple bypassing utilizing a large tantalum
or electrolytic capacitor (10 ìF or larger) which is used to
absorb low frequency variations and a small ceramic capacitor
(0.1 ìF) to prevent any high frequency feedback through
the power supply lines."
We are talking about the 0.1uF ceramic right?
also in the datasheet: (quote)
"It is advisable to keep the output compensation
components and the 0.1 ìF supply decoupling capacitors as
close as possible to the LM3886 to reduce the effects of PCB
trace resistance and inductance."
This indicates a 0.1uF soldered directly to he chips ppins would be good. I think this was suggested earlier in the thread?
Doses it have to be a ceramic or are there other types of cap which would be better?
Mike.
0.1uF is good. I would use a polypropylene over a ceramic - it's not going to be big or expensive. I'm sure i picked up some wima polyprops from maplin for about 15p each
.1uF = 100nF = 100000pF.
Monolythic ceramic is fine for power supply bypass>http://www.vishay.com/docs/45175/45175.pdf
Cheap but effective.
Monolythic ceramic is fine for power supply bypass>http://www.vishay.com/docs/45175/45175.pdf
Cheap but effective.
.
I call a vote
Local decoupling 'bypass' capacitor?
Ceramic? 1 vote so far.
Wima poly? 1 vote so far.
Or Other?
Mike
I call a vote Local decoupling 'bypass' capacitor?
Ceramic? 1 vote so far.
Wima poly? 1 vote so far.
Or Other?
Mike
mikesnowdon said:.
Local decoupling 'bypass' capacitor?
Well, if you go to Digikey, and do a search under "capacitor" and look in the features section of the different types, ceramic lists "bypass, decoupling" whereas poly film does not. Features being the recommended use for the cap in question.
Attachments
- Status
- Not open for further replies.