Jump to content

i'm on a roll... the kgsshv


kevin gilmore

Recommended Posts

So I'm getting close to the point of turning this thing on but I have some confusion about my grounding scheme. I am doing a two box on board heat sink build with the amp boards in one chassis and the ps in the other. Obviously in the ps chassis I have ac mains earth ground connected to chassis. I then have an umbilical running between the two chassis carrying both power supply ground and chassis ground, and signal ground also enters the amp chassis. What is the best way to tie these grounds together for safety and avoiding hum? I was thinking I would tie the amp chassis ground to the power supply chassis ground (and thus earth ground) and then connect power supply ground to chassis ground through a resistor inside the amplifier chassis. Does this sound like a safe scheme, and also is it then appropriate to keep signal ground isolated from chassis?

I posed the same question about signal ground isolation. I was told to tie the board grounds to the chassis ground

using 1K ohm resistors. My KGSSHV is a single box build. No noise at all be it high or low volume, music playing or not.

Link to comment
Share on other sites

I posed the same question about signal ground isolation. I was told to tie the board grounds to the chassis ground

using 1K ohm resistors. My KGSSHV is a single box build. No noise at all be it high or low volume, music playing or not.

Thanks for the response livewire. Just to clarify this means that you did not connect the signal ground?

Link to comment
Share on other sites

Sorry for not being more clear.

I did connect the the signal grounds to the chassis.

To expound on the subject, I used XLR balanced signal input jacks on the back panel,

the signal grounds from them were hooked directly to the chassis using short (<2 inch) wires

soldered to a header strip that is bolted to the rear of the chassis. The board grounds are also soldered to

the same header strip (through 1K resistors) in a centralized "star ground" configuration.

FWIW, I believe that multiple ground attachment points to the chassis should be avoided.

That would be inviting possible noise pick up into the system.

One other minor detail, if your chassis is anodized or painted, abraid the ground contact point

for the fastener to ensure a good connection directly to the bare metal.

I also use star washers to give the physical connection "a good bite".

Lastly, I use a nylok aircraft type nut to keep things from ever coming loose. A solid ground connection is what you want.

Edited by livewire
Link to comment
Share on other sites

I would try and omit the ground between the attenuator board and the amp. The amp should be happy with a true differential input.

I'd add a ground loop breaker between the PSU ground and earth. I use a pair of anti-parallel diodes, but a resistor works too.

Thanks Marc,

So I'll tie the ground from the attenuator (and hence the input ground) to the chassis ground, instead of the GND point on the amp boards. The PS ground and L/R amp board grounds will all be tied to the star ground. I'll also try placing a pair of anti-parallel diodes between the star ground and the chassis ground. Would something like this work for the diodes?

Edited by Horio
Link to comment
Share on other sites

Horio, I took the liberty of using your diagram, modifying to show what I did.

Please dont take this as "only" way, or the "correct" way.

All I know is that it worked for me.

The ground wire with the 1K resistor that leads from the psu pcb to the star chassis ground connection

is soldered directly to an unused ground pad on the psu pcb.

Also the 15 volt grounds on all three boards all tie to their respective screw terminals.

I oversimplified on the drawing - all ground planes on the three boards tie together anyway.

kgsshv-ground.jpg

Link to comment
Share on other sites

Someone needs to check the layout of the ixys current source parts on both kgsshv amp boards, andy

is having trouble, and i think there may be a problem. I'm moving one of my labs today, so i don't

have much time.

Edit: the board sure looks right, the drain connects to B+.

Edited by kevin gilmore
Link to comment
Share on other sites

The schematic does not match the board on the IXYS current source, but Kevin thinks the board is right. After we fixed the wrong facing transistor, and proceeded to replace all active parts on one board to make sure that was not the problem, there is still something wrong. When we turn the amp on, the voltage is pegged at -940 (probes in O +/-) and the balance pot will not adjust it. If it is left on more than a few seconds, one of the resistors near the balance pot starts to smoke. We are wondering if there is not a problem with the board, but we can't do any searching since we can't leave it on.

Funny thing about the 1486 being backwards, we could adjust the voltages and get them pretty good, but they fluctuated. But when we had it on with the 1486 the wrong way there was almost no heat.

Anyone have the amp boards working with the IXYS part?

Link to comment
Share on other sites

What doesn't match? From I can see drain is tied to V+, gate is tied to the 1k resistor and source is tied to the 1.2 k resistor, both of which are then tied to the collector of the sc4686, which is what the schematic shows.

Are you looking at the layout, which is correct, or the schematic. I don't know how to read this stuff myself, going by someone else's opinion. It's possible we are wrong.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.