1U NewYorkDave MILA completed

This is a rebuild of my original MILA, which was in a 2U rack and had the world's worst layout.

Front Panel








In action:






Gut shots:






Sounds awesome and works great except for two things:

1) Phantom power is giving me only 32.5V. I lifted the phantom power circuit directly from the green pre PSU. I'm using a 15V / 12VA Triad flatpack transformer for 12VDC filaments and phantom power. I've checked the wiring and circuit about 10 times and can't see any obvious mistakes. My two cheapo condenser mics seem to work fine at 32.5V though.

2) The unit is rock solid quiet when the cover isn't on. When I place the cover on the top and power it, the entire unit starts to hum/vibrate slightly. This translates a barely audible hum in any recorded track, which is not present when the unit is run with the cover off.

I thought maybe the PTs were touching the cover of the rack, so I put in some washers between the rack walls and the cover to offset it higher... but that hasn't helped. I don't *think* anything is touching but I can't be sure since I can't look inside the case when it is all buttoned up. My cheapo solution so far has been to seat the unit in the highest slot in my rack with the cover off (you can't tell by looking at it in this position that the cover is off), but obviously I'd like to be able to use it noise-free with a cover in place. Any ideas WTF is going on here?

Thanks for taking the time to look.

Mason

Nice build.

2) The unit is rock solid quiet when the cover isn't on. When I place the cover on the top and power it, the entire unit starts to hum/vibrate slightly. This translates a barely audible hum in any recorded track, which is not present when the unit is run with the cover off.

Click to expand...

Are the Triad transformers' lams grounded? I have found that EMI can circulate round cases in some instances. Also, how is your grounding scheme? Make sure the panels are not insulated from each other.

rodabod said:

Nice build.

2) The unit is rock solid quiet when the cover isn't on. When I place the cover on the top and power it, the entire unit starts to hum/vibrate slightly. This translates a barely audible hum in any recorded track, which is not present when the unit is run with the cover off.

Click to expand...

Are the Triad transformers' lams grounded? I have found that EMI can circulate round cases in some instances. Also, how is your grounding scheme? Make sure the panels are not insulated from each other.

Click to expand...

Yeah, I would say that it does sound like ground hum. I just don't understand why it only happens when the cover is on.

The 90% of the circuit is set up with a star ground (you can see it in the gut shot right next to the XLR jack. The cathodes of the tube gain stages are run to their respective filter caps, which are then run to the star. The Input and Output transformers, as well as the input and output jacks are also run to the star. The jacks are isolated from the panel with shoulder washers.  

The only part of the circuit that isn't grounded to the star is the PSU. For the B+, the diode bridge and the first two filter caps are daisy-chained and the ground runs from the 2nd filter cap to the safety ground of the IEC mains. For the filaments/phantom power, the diode bridge is grounded to the safety ground of the IEC mains.

I'm not sure if the lams of the flatpacks are grounded. I also don't know what you mean by "make sure the panels are not insulated from each other". Do you mean the top and bottom panels of the rack? They are not insulated from each other if that is what you mean. The entire rack is connected together by metal screws.

Mason

Never mind the phantom power question, I figured it out.

Very nice build!!!  Who did the front panel?
Best,
Bruno2000

bruno2000 said:

Very nice build!!!  Who did the front panel?
Best,
Bruno2000

Click to expand...

I engraved it with a friend of mine from work.

I designed it in QCad, exported the dxf to visio (a painstacking and horrible process, I assure you), and then he "printed" the visio file with a laser cutter.

Here's a quote from another thread that may help you with your lid problem.

mitsos said:

Last night I had an idea.... Too much paint (fancy that!). Basically there is paint between the faceplate and the main case (where they bolt together), so the faceplate itself is isolated from the rest of the (grounded) case. Scrape that paint off (use a knife, paint is pretty thick, one thing our friends did right!) and it's gone.

Click to expand...

dustbro said:

Here's a quote from another thread that may help you with your lid problem.

mitsos said:

Last night I had an idea.... Too much paint (fancy that!). Basically there is paint between the faceplate and the main case (where they bolt together), so the faceplate itself is isolated from the rest of the (grounded) case. Scrape that paint off (use a knife, paint is pretty thick, one thing our friends did right!) and it's gone.

Click to expand...

Click to expand...

haha... you'd think I would have thought of that, considering that I used a dremel to remove the paint on the chassis at the star ground point and at the IEC safety ground point to ensure good contact with the metal case.

MasonAtom said:

dustbro said:

Here's a quote from another thread that may help you with your lid problem.

mitsos said:

Last night I had an idea.... Too much paint (fancy that!). Basically there is paint between the faceplate and the main case (where they bolt together), so the faceplate itself is isolated from the rest of the (grounded) case. Scrape that paint off (use a knife, paint is pretty thick, one thing our friends did right!) and it's gone.

Click to expand...

Click to expand...

haha... you'd think I would have thought of that, considering that I used a dremel to remove the paint on the chassis at the star ground point and at the IEC safety ground point to ensure good contact with the metal case.

Click to expand...

Yep, this is what I was suggesting regarding panel insulation.

It's a tricky situation, a bit like quantum physics since you can't look inside the box to move things and see what's going on when the problem exists!

Looks great!  nice job.

I especially like the front panel window for that 'fresh tubes in the oven view'   

Thumbs

rodabod said:

MasonAtom said:

dustbro said:

Here's a quote from another thread that may help you with your lid problem.

mitsos said:

Last night I had an idea.... Too much paint (fancy that!). Basically there is paint between the faceplate and the main case (where they bolt together), so the faceplate itself is isolated from the rest of the (grounded) case. Scrape that paint off (use a knife, paint is pretty thick, one thing our friends did right!) and it's gone.

Click to expand...

Click to expand...

haha... you'd think I would have thought of that, considering that I used a dremel to remove the paint on the chassis at the star ground point and at the IEC safety ground point to ensure good contact with the metal case.

Click to expand...

Yep, this is what I was suggesting regarding panel insulation.

It's a tricky situation, a bit like quantum physics since you can't look inside the box to move things and see what's going on when the problem exists!

Click to expand...

Was your problem related to the lid or the frontpanel? I couldn't quite tell in your thread.

It was a real pain in the ass to get the frontpanel mounted and I'd rather not have to go through taking it off and re-mounting it again... Should I just dremel a spot on the lid and on one of the panel sidewalls to ensure contact there?

Thanks for the help everyone.

mason

Good looking build! I hope it's just the camera, but those blue lasers burnnnsss usssss!  

As for the problems, I haven't had a single build in my life where 50hz magnetic hum from PSU transformers didn't increase with the top lid on. The lid creates a mirror of sorts for the magnetic fields, and instead of radiating away from the case they are now bouncing in there - highly unpredictable behaviour as well. Best thing, but most expensive, would be a well sealed mu-metal case around the PSU transformers. Might want to experiment with cheaper metals, transformer orientation (angles), or grain-oriented steel sheets (someone sold them on the black market and they work well).

Then there's the separate issue of non-shielded wire. Seems to me like you didn't shield anything. Those high impedance range and gain pot wirings are probably picking up all kinds of crap. So are all output and input wiring (to and from transformers). "when in doubt, shield anyway", is my paranoid rule.

Kingston said:

Good looking build! I hope it's just the camera, but those blue lasers burnnnsss usssss!  

As for the problems, I haven't had a single build in my life where 50hz magnetic hum from PSU transformers didn't increase with the top lid on. The lid creates a mirror of sorts for the magnetic fields, and instead of radiating away from the case they are now bouncing in there - highly unpredictable behaviour as well. Best thing, but most expensive, would be a well sealed mu-metal case around the PSU transformers. Might want to experiment with cheaper metals, transformer orientation (angles), or grain-oriented steel sheets (someone sold them on the black market and they work well).

Then there's the separate issue of non-shielded wire. Seems to me like you didn't shield anything. Those high impedance range and gain pot wirings are probably picking up all kinds of crap. So are all output and input wiring (to and from transformers). "when in doubt, shield anyway", is my paranoid rule.

Click to expand...

The lead to the gain pot is shielded, as is the lead to the grid of the input tube. The range I didn't shield, didn't really think of it as high impedance... I've always thought of "high impedance" in the tube sense as being in 250k+ range... guess I'm wrong.

The thing is, the audio circuit itself doesn't really seem to pick up that much of the noise that is emanating from the case. It is certainly a little noisier than it was without the lid on, but not by a much. The more annoying part is hearing the case itself hum and having to listen to my wife ask "Is that thing going to blow up?". You can actually feel a slight vibration if you hold onto the power switch.

oh, and btw - the blue isn't nearly as intense as in the photos - that is more of a function of my suckiness as a photographer. I did have to increase the dropping resistor to the pilot LED though, because that was pretty damn bright at first.

mason

MasonAtom said:

The more annoying part is hearing the case itself hum and having to listen to my wife ask "Is that thing going to blow up?". You can actually feel a slight vibration if you hold onto the power switch.

Click to expand...

oh. That just means you are using the transformers (or at least one of them) near their rated maximum specs. They start to sing. I don't know your exact specs but I expect there should be "plenty more" juice available still. Just that you can't trust most transformer manufacturers. Only the really good ones give you ample headroom, and you can safely trust the spec. You can only really avoid all this by heavily over specifying your transformers. Take the values you need and double or triple them, and get that transformer. Especially true with heavy currents like heaters. If you need 6,3V 1A, give it a 2A (or so) transformer.

But you were probably limited with transformer choices when designing for the 1U unit.

MasonAtom said:

Was your problem related to the lid or the frontpanel? I couldn't quite tell in your thread.

Click to expand...

What was going on with my (chinese) neve clone is that the front panel was isolated from the rest of the (grounded) case. It was somehow acting as an antenna and inducing noise (into the gain switch?) when the shaft of the gain switch touched the inside of the panel hole. I could induce this by bringing my hand close to the gain switch while the thing was working. All was gone as soon when I de-isolated (and hence grounded) the front panel along with the case.

good luck with your issue. looks mighty nice. can't wait to get mine done!

Mason - nice, tidy work!  If it sounds as good as it looks you're in business.

Kingston said:

... I haven't had a single build in my life where 50hz magnetic hum from PSU transformers didn't increase with the top lid on. The lid creates a mirror of sorts for the magnetic fields, and instead of radiating away from the case they are now bouncing in there ...

Click to expand...

That's interesting.  I lot of old gear had lids that were more of a screen or a metal grid.  I wonder if that would fare better than solid sheet metal.  I made a tube amp kit and used a metal screen for the lid.  For sure it helps let the heat out, but the amp does get dusty inside.

Really good looking !

Ian Taylor said:

Mason - nice, tidy work!  If it sounds as good as it looks you're in business.

Kingston said:

... I haven't had a single build in my life where 50hz magnetic hum from PSU transformers didn't increase with the top lid on. The lid creates a mirror of sorts for the magnetic fields, and instead of radiating away from the case they are now bouncing in there ...

Click to expand...

That's interesting.  I lot of old gear had lids that were more of a screen or a metal grid.  I wonder if that would fare better than solid sheet metal.  I made a tube amp kit and used a metal screen for the lid.  For sure it helps let the heat out, but the amp does get dusty inside.

Click to expand...

Well think about it in a tight studio space. When the magnetic field radiates outside the unit, where is it gonna go? First stop is your other gear, which might now hum a bit more.

No free lunch.

Best thing to do is to try to stop it at the source, in any way you can.

Update:

I used a dremel to remove some paint on the lid and panel walls at the points of attachment. This helped cut the vibrating/humming of the case by about 50%.

I'm planning on doing the same with the front panel, just haven't gotten around to it yet.

Regarding overtaxing the PTs: the filament/phantom power PT is 12VA rated for 800 mA at 15VAC. I'm reading 440 mA on the heaters at 11.94VDC, so this should be well within spec. The B+ PT is rated for 50 mA at 230VAC. The circuit only pulls about 11 mA, so I should have plenty of headroom there.

Mason

Yeah, with the Par-Metal chassis I shave off the powder coating at all the corner screwholes on every panel, including the faceplate.  This effectively connects the entire case, and when grounded results in nicely quiet units. 

Very nice work.  Looks great!

JC