Another BeePre 2 with Hum

[Paul Birkeland]

Perhaps just a bad component?

No, that's not how this works unfortunately. 

A resistor will smoke and blow up because you drew excessive current through it. I have circled in green one possible cause.  If the lead circled and the middle terminal strip lug circled touch each other, you'll send that power supply dropping resistor up in smoke. 

The leads circled in red should be bent straight up so there's no risk of them touching the chassis. 

[ScottAstroNut]

Thanks, Paul,

It's possible that there could have been a short where you have circled the green, and I had thought of that, too, but upon close inspection I saw a gap between the grounded terminal and the lead off the 22 ohm resistor. I also didn't see a spark, just saw smoke.  But again, I could be mistaken.

Regardless, the 270 ohm resistor is toast.

Will do on the other circled red items. That was my mistake. Being careless.

What would you recommend at this point? Make the adjustments you suggested and replace the 270 ohm resistor?

[Paul Birkeland]

I would make your amp look as much like the one in the manual as possible.  I wouldn't leave that resistor lead like you have it, that may continue to smoke your power supply randomly if you don't address it. 

I would replace the 270 ohm 5W resistor after you've made the adjustments and give it another shot.

[ScottAstroNut]

Will do.

Do you recommend I contact Eileen for a replacement or simply purchase one from one of the electrical supply companies?

[Paul Birkeland]

Whichever one is more convenient for you.  For a single part order, it tends to be far less expensive to just use us.

[ScottAstroNut]

Great. Will contact Eileen.

Do you think that the proximity of the 22 ohm wire to that ground lug could have been a source of the hum?

Thanks for your help!

[Paul Birkeland]

I don't really know, it's certainly worth the effort to correct that area and see if that helps.

[Paul Birkeland]

I was mulling this over while doing some other work, and that short wouldn't blow out that resistor, but shorting that connection to ground will remove one of the filtering elements in the power supply and increase the noise floor.  This wouldn't be a huge increase in noise, but it would be audible, and it would come with ever so slightly higher than expected voltages. 

[ScottAstroNut]

Paul, I was thinking about that, too. The combined resistor/capacitor/diode assembly is essentially both a high pass/low pass filter, isn't it? So in one sense it is already a short to ground as seen at terminal 36? Shorting the 22 ohm resistor to terminal 35 therefore shouldn't be able to take out the 270 ohm resistor between terminals 36 and 39. In which case, what did? All I did was turn the power on to a circuit that was already working, albeit with a slight hum.

Also, the readings I measured at terminals 6 and 16 in the last voltage test measured lower than the expected 100VDC by 7%, not higher.

What I can say is that the oscilloscope trace of the output with no input signal was not clean. My friend the audio engineer helped me with the oscilloscope readings. What we saw was baffling enough to him that he wanted to trace the signals through the circuit. That was what we were getting ready to do when the 270 ohm resistor blew. We hadn't touched anything in the circuit, however, before the resistor went up in smoke, so it wasn't caused by a slipped probe. All I had done was invert the circuit in the building jig I use during construction and testing. Rather than use the wooden base provided as a building support, my jig allows me to invert the circuit with tubes and transformer attached without bottoming out. Just more convenient and secure.

[Paul Birkeland]

If you flip the amp over and a resistor goes up in smoke, something may have been really close to touching where it shouldn't have, or barely touching but then touching too much when you flipped it over, so in a way you have a more solid hint at what's going. 

I repaired an old Pilot EL84 amp that relied on ground contacts built into the 9 pin sockets.  It would only work upside-down, when you flipped it back over the sockets were loose enough that those connections would open up and the amp would do strange things.  This isn't all that related to what you're experiencing, but similar in the sense that you have a hint now that something may be moving around more than it should.

[ScottAstroNut]

My next question, then, is how to go about diagnosing that without blowing another resistor once I get a replacement and have repaired the circuit?

As I mentioned, at one time I did electrical engineering, but it was all digital logic systems. I did my circuit design on breadboards or perfboards with lots of wire wrap. If I mis-wired something, usually it would not go up in smoke! I did learn about regulated power supplies, but only to the extent that I could build one that would provide the 6 to 12 VDC I needed for my digital IC's. You analog (and especially you audio tube-analog!) guys work in an entirely different world. One that I am very much enjoying getting to know a bit better by building your kits. I have a lot of respect for the expertise and experience you need to do your kind of job.

Having circuits that work only upside down or right side up is amusing to me! I love EE war stories!

[Doc B.]

I think you have to look at this from two different perspectives. Yes, you can analyze the circuit and peruse the schematic to see if you can imagine a scenario where something that isn't exactly like the circuit drawing is causing the issue. But from a practical standpoint it's a lot easier to say "crap, something shorted" and look at the amp for signs. There are various ways this can happen. Gravity - when you flip the amp over something moves and shorts out. Gravity, part 2 - when you flip the amp over an offcut piece of wire that was hiding somewhere falls down and shorts something. Time - there was a miswire from the beginning that finally made the resistor give out from being overloaded.

If I was stuck I would measure across each filter cap and look for a low resistance reading to ground. If I found one I would start to investigate the components nearby. When a big resistor smokes there is a short somewhere. Miswire, leads touching things they shouldn't, caps/diodes in backwards - the list of possibilities is not too long.

[ScottAstroNut]

Hi, Doc! Thanks for the helpful insights.

I did check the caps in the HV supply, and did find that the 220 microfarad cap closest to the transformer has a very low resistance reading. So I guess I am going to need a new capacitor, then. I don't immediately see anything amiss in the vicinity of the cap, however. The diodes look to be OK. None appear to be oriented the wrong way. The other three 270 ohm  resistors don't appear damaged when measured. The other two large caps seem fine.

[Paul Birkeland]

I did check the caps in the HV supply, and did find that the 220 microfarad cap closest to the transformer has a very low resistance reading. So I guess I am going to need a new capacitor, then.

No, lift one leg out of the circuit then check again.  If the cap is still soldered into the amp, you can't say much about its condition.  The caps we provide are of exceptional quality, so it would be surprising to randomly get a dead one. 

[Deke609]

That seems like a really weird place for a resistor to fail: the last resistor in the return (-ve) rail of the B+ supply. Granted, I don't know much about this stuff, but I can't think of a way that could happen that doesn't involve an obvious serious wiring error - and I don't see one in your photo.

So a shot in the dark: how much physical tension do you have on those white cement capacitors? I can imagine that the internal connection between a lead and the internal wire coil (or whatever resistive structure is inside) might crack if you soldered one lead and then pulled the other really tightly before soldering it -- or installed the resistor suspended by the leads and then tried to push the resistor towards the chassis.  A cracked/flaky connection inside the resistor might easily overheat with even normal current running through it and then pop/smoke when it inevitably fails. I can also imagine it being noisy when it was still capable of passing current.  Just a thought. 

cheers, Derek