[Emon]

I built a Millet Hybrid and a STEPS over a month ago. It worked great. But it was sitting in a cardboard box because I never finished building the chassis. Well, I did that a few weeks ago, and at the same time made all the connections modular so I could mount all the components easier. The biggest change is that I had my Alps RK27 on a small piece of perf board with a 10 pin ribbon cable leading back to the PCB. I checked and checked again, and all the connections were fine.

But when I turned on the amp without a source, a 60-ish Hz hum buzzed loudly in both channels. Maybe more than 60 Hz, sounded like a fluorescant light hum. But when I plugged in my source the sound almost went away, only audible with the pot at the high end (I run it at about 9 o'clock with my E-MU 1212m). I just finished putting it together so I wasn't going to bother with it, especially since it wasn't bothering me much.

So today I decided to figure out what was going wrong...I figured just a simple grounding issue or something. I removed the ribbon cable setup and used some better wire (I think the ribbon cable connector was shoddy), hardwired onto the pins of the pot. Well, the hum was still there. So I fiddled around in my STEPS a little, thinking maybe I accidentally fried my LM317 with static when I was carelessly moving the boards around when putting them into my chassis and modularizing all the connections. I thought maybe it wasn't working right and letting real noisy power into the amp or something. So I replaced it, the hum was still there. Then I accidentally shorted one of the power diodes (one of the set of four in the TO-220 packages) to ground.

Next time I'll put heatshrink over those. I measured the voltage drop on the supposed dead diode and the reading was the same (or close to) the other diodes. I waited a while, plugged in the AC and got arcing under the AC line filter. I'm thinking since I fried one or more of the semiconductor components the AC arced back to ground since it had no where else to go.

So my question is: What could have caused the hum on my Millet Hybrid, and what's dead on my STEPS? I was rather careless when modifying my Millet and STEPS, letting the bottom of the boards slide around my old, almost fuzzy workbench (it's some crappy particle board). So I'm thinking I fried one or more semiconductor components on the Millet, probably the buffers. For the STEPS I probably cooked the LM317 and one or more of the diodes.

Also, could any of the box caps be damaged despite the complete lack of physical evidence? I've seen plenty of electrolytics pop, but never a box or tantalum cap. There's no melting, no soot or scorching under the caps, so I'm thinking they're fine.

So any thoughts? Right now my plan is to replace all the semiconductor parts on both my Millet and my STEPS. Shouldn't be more than 15 dollars I think. Are there any other components that could be dead? I'll also probably go ahead and order some 12FK6 tubes to replace my 12AE6A tubes (which aren't even matched). But I doubt the tubes had anything to do with it.

 

[n_maher]

The only real advice that I can give is about the grounding hum that you were experiencing with the Millett. I recently built an amp very similar to the one you're describing and had a heck of a time getting all the hum out of it. Here's what it took for me:

1. Route all of the signal wires as far as humanly possible from the STEPS while keeping cable lengths to a minimum.

2. Make the cable between the STEPS and the amp board as short as possible.

3. Mount the steps as far away from everything else as possible.

4. Ground everything appropriate to a single, central point.

That removed 95% of the hum that I had. The other 5% all but disappeared when the amp was plugged into a source. I also have about the world's worst power in my house (200 year old) which I'm sure doesn't help.

As far as your other questions I'm sorry to say that I don't have much to offer.

HTH,

Nate

 

[amb]

Quote:

Originally Posted by n_maher Make the cable between the STEPS and the amp board as short as possible.

This is not really that important. In fact I have the STEPS in a separate chassis than the Millett, connected by a 3-foot long umbilical, and there is no hum at all. The umbilical is a home-made cable with two 18-gauge stranded wires twisted tightly together, wrapped in techflex, and terminated on both ends with 2.5mm barrel-type power plugs.

Keep the power transformer as far away from the amp board and input wiring as possible, and ground the amp case and volume pot body.

 

[n_maher]

Quote:

Originally Posted by amb This is not really that important. In fact I have the STEPS in a separate chassis than the Millett, connected by a 3-foot long umbilical, and there is no hum at all. The umbilical is a home-made cable with two 18-gauge stranded wires twisted tightly together, wrapped in techflex, and terminated on both ends with 2.5mm barrel-type power plugs.

I was just trying to convey the idea that any wire inside the chassis (assuming both the PS and amp are in the same enclosure) can act as an antenna and pick up noise, especially one so close to the transformer. Wouldn't you agree that your Millett doesn't have any PS hum due to the fact that you put the STEPS in a separate enclosure? I think that once you move the STEPS out of the same box as the amp that power chord length is not an issue.

And just like that, I've posted a thousand times...

Nate

 

[amb]

Quote:

Originally Posted by n_maher I was just trying to convey the idea that any wire inside the chassis (assuming both the PS and amp are in the same enclosure) can act as an antenna and pick up noise, especially one so close to the transformer. Wouldn't you agree that your Millett doesn't have any PS hum due to the fact that you put the STEPS in a separate enclosure? I think that once you move the STEPS out of the same box as the amp that power chord length is not an issue.

The voltage regulator has low enough output impedance that noise induction into the DC power wiring shouldn't really be of concern, whether the PSU is in the same case as the amp or not.

Quote:

And just like that, I've posted a thousand times...

Congrats!

 

[mono]

Quote:

Originally Posted by Emon But when I turned on the amp without a source, a 60-ish Hz hum buzzed loudly in both channels.

Are you certain you had all diodes oriented correctly? Maybe one or more are backwards?

Quote:

Then I accidentally shorted one of the power diodes (one of the set of four in the TO-220 packages) to ground. Next time I'll put heatshrink over those.

Maybe heatshrink over the part you shorted them with too (for example, probes?).

Quote:

I measured the voltage drop on the supposed dead diode and the reading was the same (or close to) the other diodes.

Check it the other direction too, the diode check function on a DMM.

Quote:

I waited a while, plugged in the AC and got arcing under the AC line filter. I'm thinking since I fried one or more of the semiconductor components the AC arced back to ground since it had no where else to go.

Not sure what you mean by "nowhere else to go". Power is not compelled to go anywhere, it'll terminate fine anywhere there's an open circuit. The arching would be a component failure or short most likely. What does "under AC line filter" mean exactly? On the component itself or on bottom of circuit board or? If on bottom of circuit board I'd suspect you had the board on a conductive surface. If on the filter itself I'd suspect either the windings enamel is scratched off from mechanical damage or it'd shorted previously and the heat burnt off some enamel- it'd probably look a bit discolored previously if that were the case.

Quote:

So my question is: What could have caused the hum on my Millet Hybrid, and what's dead on my STEPS? I was rather careless when modifying my Millet and STEPS, letting the bottom of the boards slide around my old, almost fuzzy workbench (it's some crappy particle board). So I'm thinking I fried one or more semiconductor components on the Millet, probably the buffers. For the STEPS I probably cooked the LM317 and one or more of the diodes.

Test each in isolation. Get the steps working before fiddling with the MH, or use another supply in the interim for the MH. I would not assume you cooked the LM317 yet, but before installing a different LM317 you can just leave the old one out of the board and measure voltages- the STEPS will be electrically functional enough for you to measure input voltage (the LM317 "input" pin) with a DMM before installing a new one... or even before removing the old one but with the arching you should find that cause before reapplying power to *any* of it.

Quote:

Also, could any of the box caps be damaged despite the complete lack of physical evidence? I've seen plenty of electrolytics pop, but never a box or tantalum cap. There's no melting, no soot or scorching under the caps, so I'm thinking they're fine. So any thoughts? Right now my plan is to replace all the semiconductor parts on both my Millet and my STEPS. Shouldn't be more than 15 dollars I think. Are there any other components that could be dead? I'll also probably go ahead and order some 12FK6 tubes to replace my 12AE6A tubes (which aren't even matched). But I doubt the tubes had anything to do with it.

The caps are probably fine but you could check for resistance across them, remembering that you will get a high ohm reading because of the resistor. They wouldn't generally fail from (whatever happened) though and since they dont' even have polarity, should be fine.

Trace the failed part (or misinstallation) before trying to randomly replace parts. After carefull examination of the board, and being mindful of the dangerous high voltages, DMM probes can take voltage readings anywhere, literally, just be sure not to short anything, take your time and proceed with caution first testing with no power applied. Above all, be mindful of safety when working with line voltages.

 

[Emon]

Quote:

Originally Posted by mono Are you certain you had all diodes oriented correctly? Maybe one or more are backwards?

I'm sure. It used to work perfectly. I didn't change anything on the STEPS.


Quote:

Originally Posted by mono The arching would be a component failure or short most likely. What does "under AC line filter" mean exactly? On the component itself or on bottom of circuit board or? If on bottom of circuit board I'd suspect you had the board on a conductive surface. If on the filter itself I'd suspect either the windings enamel is scratched off from mechanical damage or it'd shorted previously and the heat burnt off some enamel- it'd probably look a bit discolored previously if that were the case.

By "under the AC line filter" I mean between the PCB and the casing of the line filter. The pins are too thick to enter the PCB holes for some 1/8", so it's offset from the PCB a little. The windings on the topside all looked fine. No scorch marks or anything. I didn't really look under it though.

Anyways, I'm not going to have time fix this before I go back to school. I really do like my E-MU 1212m + ER-6 combo, that's good enough for now. Plus I'm in temporary housing for god knows how long, I'm not taking expensive cans with me just yet.

Thanks for the info though. I guess I'll look at it over fall break in a few months.