[TrevorNetwork]

Hi all,

I have cased up the dynahi (similar to Rsaavedra's) and I now have a hum (60hz). Unfortunately it is induced on the board, and not the input wires. So shielding the input wires will not really help. I removed the input wires from the board all together, and still heard a light hum. I have tried rotating the power supplies (Power Ones) to no avail. This issue did not occur prior to casing it up. The person that made my enclosure it willing to make me a solid piece of 1/4" aluminum with a 3/4" hole near the bottom with a grommet to feed the wires through. I could place this wall in the centre, to divide the psus from the pcbs. Would this help at all? Or is my only solution to separate the psus from the pcbs, and run an umbilical?

 

[Megaptera]

What kind of transformer is it? I've been reading up on power supplies lately for my own project, and from what I've read, it seems that the ways to reduce the interference are to shield the transformer, use a toroidal, and keep the transformer as far from the board as possible.

 

[TrevorNetwork]

It seems to be a standard transformer:

http://www6.head-fi.org/forums/showt...2&page=3&pp=25

It is most certainly not a torroid. It is not easily replaceable either.

 

[rsaavedra]

You know Trevor reading my "Sound Reinforcement Handbook", page 283, I find the following:

Quote:

"Cable shielding is primarily for electrostatic noise (...). But there is another type of noise. Electromagnetic noise (...). Normal cable shielding does not exclude electromagnetic fields (unless the shield is heavy iron or steel conduit). Magnetic fields are cancelled only by a balanced line, with twisted center conductors, and by sheer physical distance from the source"

Starting to think an aluminum wall + heatsinks are possibly doing just electrostatic shielding, but not electromagnetic shielding from those strong transformers' fields. I'm leaning towards replacing the aluminum wall I have with either a thick iron or steel wall, or maybe several layers of thin iron plates to make a thick iron wall. Haven't heard my amp yet, but if you are getting hum from proximity between PSU's and boards my guess is my amp will probably have some audible hum too.

 

[TrevorNetwork]

I am simply going to have another case made, just for the PSUs. I am going to use an umbilical cable from the PSU case using an XLR male, and an XLR female on the amp chasis.

 

[peranders]

If you use a toroid the problem is your grounding I think. Have you used the "star grounding concept"?

My QRV-07 amp has tow non-toroids very close, 2" away and I have no hum at all.

 

[rsaavedra]

Quote:

Originally Posted by TrevorNetwork I am simply going to have another case made, just for the PSUs. I am going to use an umbilical cable from the PSU case using an XLR male, and an XLR female on the amp chasis.

That's the safest bet. I'll try the iron wall, if I end up with hum I'll have to consider psu's in another case too.

 

[peranders]

Quote:

Originally Posted by TrevorNetwork I am simply going to have another case made, just for the PSUs. I am going to use an umbilical cable from the PSU case using an XLR male, and an XLR female on the amp chasis.

Before you do too much just make the cables longer and see if this really has any effect. If you have done it right you can have the toroid one inch or two from the pcb without any hum. So... check your grounding.

What happens if you disconnect one channel? Still humming?

 

[TrevorNetwork]

The transformers in question are not torroidal.

 

[peranders]

In that case you might have a little bit bigger problem but check first before you build any cases.

 

[rsaavedra]

Quote:

Originally Posted by peranders In that case you might have a little bit bigger problem but check first before you build any cases.

Peranders, not only his transformers were not toroidal, he didn't have any humming when the psu's were far from the pcb boards of the Dynahi. It seems only when casing everything together he gets the hum, so proximity between psu's and main boards seems to be the culprit.

 

[rsaavedra]

Went to Lowe's and saw some 12"x24" zinc plated steel sheets (26 ga), just $4, and thought of something that might sound ridiculous, but I think might be a reasonable idea. Besides a thick steel wall between PSU's and boards, how about also shielding the boards all around? Kind having them in their own steel box about 11" x 5" x 2.5" inside the chassis. Won't really "box" them, but will have a 5"-wide stripe of steel sheet go all around the boards, basically clothing them from the edges of the heatsinks all the way back. Actually, could even close the back (really "boxing" the boards), with holes just for the input signal wires, output wires, and power wires.

 

[gsferrari]

humming can also arise if the inputs and ouputs are not grounded. Inputs grounded to board and outputs to the chassis. Dont ground the pot/attenuator to the chassis - straight to the board.

I dont think the humm is because of the transformer or power supply...shielding will not help in this case because you will connect shield to chassis and this is again the ground. You dont want to share this ground with your inputs and outputs and headphone outputs...attenuator etc. if it is used as shielding.

I think this is grounding related and a solution can be found without moving the power supplies to a separate chassis.

 

[rsaavedra]

Quote:

Originally Posted by gsferrari Dont ground the pot/attenuator to the chassis - straight to the board.

Actually Kevin recommended star grounding, so better to have a wire from the attenuator's ground to the star ground point, not to the board's ground.

 

[morsel]

Before you go crazy on mu metal shielding, make sure you have followed proper grounding technique. Does hum level change with volume knob position, if you touch signal ground on the amp, or disconnect the amp from the source?