1st project hum or buzz

[Carlp]

OK, I'm back after a little tinkering with my first SET (Fred Nachbaur's Miniblok, http://www.dogstar.dantimax.dk/tubestuf/miniblk2.htm)... well, 1st amp of any kind really, and I'm new to electronics though I'm a quick study.

It isn't pretty as I was figuring out some of the layout as I went, but in the end the amp has a hum or buzz through the speaker and I'm looking for troubleshooting thoughts before I start digging.

Here's what I know. I followed Fred's design carefully and most of my initial testing was spot on (once I remembered that the socket pin layout is reversed from the top vs bottom! Arrrggghhh). Only exception is on pin 1, which he has at -30V (neg. grid bias) and my Fluke meter seemed to indicate only -1.5V. Also, he includes an LED in the B+ voltage that indicates capacitor charging and filament heating (I think...). Although my LED shows the 300V B+ on the + side, it doesn't light at all. I haven't dug into that problem but all wiring and connections seem to be OK.

That's it. Oh, and my layout. Power comes into the back, not far from the signal input (it's only 2x4" back there) and I used an unshielded interconnect for that (Fred suggests shielded). And I have a few longish runs of signal wire to the pot and tube in a small housing (2x4x6"). I disconnected the interconnect cable but the hum or buzz remains.

OK. I should say I was quite impressed by both the sound quality and volume (through some old Ohm speakers, not especially efficient if I recall). Hard to tell with only one channel and my quick test, but it's enough to get me hooked. Particularly the low cost of this design.

Thanks so much for any suggestions what to look into.
Carl

[DigitalJunkie]

Use shielded wire for your signal wiring,and re-route all of the AC wiring,so it is as far away from the signal wiring,and the tubes grid circuits as possible.

Also,You might want to double-check everything in the bias supply!
Double-check the wiring,and everything,and if needed check the individual components with a DMM. Something's fishy here.

Are the power transformers and output transformers mounted close together? If possible,mount the power (or output) transformer(s) 90-degrees to the others.

[Carlp]

Thanks, DigitalJunkie. I did mount the power and output transformers at 90 degrees (I'd stumbled on that here somewhere I think - this forum is a GREAT resource for a beginner like me). If by DMM you mean digital multi meter, that's what I used to test the amp before trying it with the tube and a speaker.

I'm curious, though, why if the power grid voltage was really so far off (-1.5V instead of -30V or more), it seems like I wouldn't have gotten much sound out of it (particularly since it's only 1 watt output max to begin with). Yet I got a surprisingly loud (and good if you ignore the hum) output.

I'll try shielded and re-routed signal lines (I did try to keep them apart but had little space to maneuver with). What should I use as shielded signal lines, and do I need to ground the shield? Coaxial cable (it'll be tough to fit in the current enclosure)? Is there some other kind of wire I could use? Any suggestions for sources of shielded wire?

Thanks again,
Carl

[Sherman]

Originally posted by Carlp
...

I'll try shielded and re-routed signal lines (I did try to keep them apart but had little space to maneuver with). What should I use as shielded signal lines, and do I need to ground the shield? Coaxial cable (it'll be tough to fit in the current enclosure)? Is there some other kind of wire I could use? Any suggestions for sources of shielded wire?

Thanks again,
Carl

Carl,
You can buy shielded wire at Radio Shack. I recommend that you use single conductor shielded wire. Use the conductor (center part of the wire) to run from the input jacks to the volume control. Ground only one end of the shield (the braided part surrounding the conductor). Usually the end by the input jack is grounded and the other end is not.

Also try insulating your speaker terminals if they aren't already. What I mean is that the negative terminal of your speakers should not be grounded to the case or connected to any other ground.

All AC wires should be twisted together and as Digital Junkie said they should be as far from the signal wires as possible.

As a last suggestion do a search here for 'star grounding' and implement it in your amp.

Together all those things should greatly reduce or elminate the hum.

[tubelab.com]

If the bias voltage is really only -1.5 volts, the output tube will be drawing far too much current. This will cause a hum and short tube life. Fix that problem first.

Check the voltage at the positive end of C1, then check the voltage at the positive end of C3. This should be about 4 volts less. If the two voltages differ by more than 5 volts you are drwaing too much current. This is caused by the missing bias voltage. It likely zapped your LED also.

[Jeb-D.]

Since your a New-B I'll give you the lowdown in regards to hum in amplifiers.

-Practical matters as far as building goes:

Biasing the Tube at a higher current then the design was meant for can cause hum.

Small audio signals should be ran in shielded cable. And all audio signals shoud not run right next to AC power lines.

Assuming the design your building has an appropriatly designed power supply (right cap & inductor values for the ammout of idle current). If you use cheap rectifier diodes I've found it can cause audible spikes in the power supply, regardless of the filtering measures taken. It is best to use soft-recovery diodes, such as HEXFRED's. Which can be purchased from www.tubesandmore.com, if you don't know of any sources.

The way you have the amp grounded is very important. There sould only be 1 path to ground/for the current to return. Keep in mind current returns to the power transformer, the ground is just for voltage reference. If not, you have whats called a ground loop which will cause a pretty loud buzz. Also, when you hook an input device to the amp, it becomes part of the circuit. This can make grounding more complicated when your using an audio signal that has a grounded output (such as a PC). But for sources that are floating, such as most CD players it's much more leanent when it comes to grounding schemes.


-Technical matters as far as design goes:

SET amps are the most likely to have an audible hum because any ripple in the power supply will make it to the output. You need extremely low ripple to not have it be audible. Also they idle at higher current(class A) . This requires a more intensive power supply(bigger cap values, chokes ect). Ripple can be easily calculated

Vripple = Idle current/(frequency x capacitor value)

Frequency is 60 for Half wave rectifiers and 120 for full wave and bridge (in the US anyhow). As you can see the higher the Idle current the more ripple.

Push pull amps, even when opperated in class A (higher idle current) are less likely to suffer from hum, because in push pull stages like-signals cancel out. Of course in real life, the cancelation effect is not %100 because the tubes, cap values and transformer windings arn't exact matches. But it still helps alot.

[Carlp]

Sherman,

Thanks for the guidance. I actually just checked my nearby Radio Shack today and they didn't have any shielded wire (other than coax, which I have a ton of - can that be used if I can fit it in my tiny chassis)? In fact, I'm not sure they even know what it is as they showed me simple insulated wire when I asked. As for the speaker terminals, I hadn't heard about that and will look into whether or not I've done that. Lastly, what do you mean by twisting the AC wires? Do you mean the hot and neutrals (correct terms here? I'm used to house wiring)? This design uses AC filament heating (I think it's AC all the way around) and in this design I'm not sure I can fully implement that, but I'll work on it.

Tubelab,

If I'm really pushing -1.5V but drawing a ton more current, will I still get the same output level out of the Transformer (remember, I got decent output sound level)? I guess all else held constant that makes sense, but I'm not sure... In any case, I'll check the voltage again as from what Jeb-D said, that could be the source of the hum (and as you say, a burned out LED).

Jeb-D (and maybe others),

Where can I learn more about grounding (aside from the star grounding Sherman suggested)? I'm not sure I fully follow what you're suggesting, but I know grounding can be crucial. I'm particularly interested in understanding the ground loop issue. (For example, what physically do you mean by "one path to ground" especially in this design - note the secondary on T1 is connected directly to ground, while the "secondary" on T2 (actually the primary winding of a "backwards" transformer) is connected to ground through a 470uF cap.) Lastly, Fred (the designer) speaks to the ripple issue in his design saying "Note that we don't bother converting to DC, in this application an AC heater is just fine because 1) we're not dealing with extremely high gain, and 2) the 13EM7's indirectly-heated cathode is naturally highly immune to AC hum." That would seem to suggest ripple isn't a likely cause, but maybe I misunderstand...

Thanks so much again, folks. You are all terrific resources.

Carl

[tubelab.com]

It is very possible to have good sound (and maybe some hum) with too much current. In fact many SE amps sound best when operated above the recommended bias current (depending on the output transformer). It causes short tube life.

[Jeb-D.]

The ripple I'm talking about is in the plate/anode power supply not being pure DC. The ripple the designer is refering to is that in some applications, heating the heating element with AC will cause a slight hum. Because some electrons from the heater will hit the plate. I have found this to be a rare circumstance. AC heating seems to cause no problems in most designs.

Regarding the grounding. Unfortunatley, most schematics don't show grounding locations/technique because it would look really clutterd if they did. "Star" and "Buss" grounding are the two common methods of grounding. What I use, is usually a mix of the two. Run a google search you should be able to come up with something usefull. If not, let me know and I'll draw up a diagram for you.

If your using Radio shack diodes for your power supply I'd strongly suggest getting some HEXFRED's. In my first SE amp I could not get rid of hum with radio shack diodes. Even using rediculously huge capacitors and inductors to filter it.

[Carlp]

As for the voltage, it now measures about -5, so there's something I need to fix, and will do so straight off.

Regarding grounding, after doing some reading here and elsewhere, it seems I inadvertently replicated nearly a star grounding arrangement, but I think I need to work on that. And better ensure the metal chassis is grounded (it's grounded via one of the components, but I'll put in a single grounding point with paint well removed). Also, I'vetied that to the "earth" leg of the incoming household supply...correct?

Finally, regarding Radio Shack diodes, while most of my components were from Antique Electronic Supply, RP Electronics, Futurelec, etc., I did rely on a few Radio Shack diodes (the 4002s, which I used 4003s for) and a few small caps. I'll replace them.

Thanks again. You all are incredibly helpful.
Carl