Finished Amp/hum issue!

[basschops1528]

Hey Guys,

I finally finished the Plexi Clone and it sounds pretty good. As you'll see in the video, there's a hum problem I need to figure out.

If you guys have any ideas please do give me your input. Also, I feel the tone knobs don't affect the sound as much as I thought. The bass isn't much bassier, the middle don't suck the mids as much as you would think, and the treble isn't as trebly either. Tell me what you think. https://www.youtube.com/watch?v=N9KUvLfaFBY&feature=youtu.be

[sluckey]

Pull the first two little tubes, leaving only the PI and power tubes in place. Still hum?

I would move the OT away from that preamp tube. Put it near the output tubes. You have your speaker wires run very close to the V1 socket. V1 is very susceptible to hum and that OT is a big hummer.

IEC to fuse holder... I would put the hot wire from the IEC to the end lug of the fuse holder, not the side lug. This is a safety thing.

[ac427v]

Lovely work, and tone...but the hum sucks. Unrelated note: is there solder on the dropping resistor coming from the screen cap?
--Craig

[basschops1528]

So in order, this is what I've done and the result.

1. Removed V1 input tube-- Hum is the same
2. Then removed V2 tone stack tube-- Hum significantly reduces to ear next to speaker level.

So V2 is probably the cause of it.

I'm going to try swapping V2 with V1 and V3 to see if it's the tube.
I will also take another close up of V2 socket wiring

[ac427v]

At 3:34 in the video. The left lead on a three watt resistor is not soldered.
--Craig

[jojokeo]

At 3:34 in the video. The left lead on a three watt resistor is not soldered.
--Craig

Good catch ac427 it does appear that there's no "sodda on that resista & cap connection". (just bustin ya chops basschops  ) Go through the entire amp's solder connections visually and touch-up anything that even looks suspect. With a hum this bad it's likely that solder non-connection or something mis-wired rather than layout or grounding scheme at this point? But we shall see...?

Also, can't see where you ran the virtual center tap ground for the heaters? And can't tell at all about your overall grounding system? Lastly, you may want to remove all the electrical tape as it will eventually fall off, zip ties are best. Good luck!

[basschops1528]

Yeah I'm pretty sure it's soldered because I yanked on everything with a stick afterwards but I will doublecheck. I think the solder just dropped down or something strange but good catch.

[Willabe]

Also, can't see where you ran the virtual center tap ground for the heaters?

Do you have a heater CT going to ground or a virtual center tap ground for the heaters?

[basschops1528]

There is a virtual ground for the filaments. Two 100ohm resistors to ground from each secondary. The actual CT is heat shrunk and attached to the bottom of the PT for safe keeping.

[eleventeen]

I too believe you are having a bad OT/V1 interaction. And as Sluckey said, could be from both above and below the chassis.

Do you have a piece of sheet metal that you can ground & insert between the OT and V1 just to quick-check?

I believe you may need to move the OT. Sorry to say it. Your hum could be incurable with the OT mounted where it is...because where the OT wires need to run relative to V1. Not saying it is or isn't. 

Taking the amp apart to where one can safely re-drill for a giant part like the OT is not necessarily easy. I think I might try unbolting the OT from the chassis and running it with 3 qty 2-foot carefully insulated long wires (caution! highest amp voltages) and a 1/4" jack to a speaker (or, a scope) to prove to myself that it was an "adjacency" problem. Attention has to be paid to how the wires to the OT are routed and you could have them in a bad spot. That could be a long topic.

My point is: IF that's your problem, very likely nothing other than moving the tranny is going to fix it. You clearly have work & $$ invested in the amp, you're in, dude.

[basschops1528]

So I managed to get myself some sheet metal but its a piece of stainless about 1.5mm thick.
I plan to try doing some test shielding as you guys have suggested but I have an idea as well.
What if I buy or build the parts for an electromagnetic pickup with a resonant frequency at 60Hz?
I could use an inductor for the pickup and a band pass circuit to limit unwanted EMI, cover it in epoxy or parafin and attach it to a noncunductive stick to find areas moat susceptible like tubes or wires. Any thoughts on that and how it could be done? Would be great to use with a scope and on any project where hum is a problem. 

[PRR]

> a piece of stainless about 1.5mm thick.
> an electromagnetic pickup with a resonant frequency at 60Hz?

I vote for "move the OT away from that preamp tube".

You have your out-house right next to the drinking-well. An extended poop-pickup test would just confirm the obvious: move the outhouse.

Shielding is rarely effective. The metal can on the first tube is not bad, but won't keep heavy crap out.

[silverfox]

Just out of curiosity, what is your bias set at? In other news, I had removed my circuit board to add a preamp to it and today got it wired back in. Before I finish connecting the preamp to the existing power amp, which was working, I decided to test the power amp to make sure I had it wired up properly. When I powered it up there was a louder than usual hum associated with the amp and the voltage wouldn't come up fully. It turns out one of my bias wires was unconnected. After I found that and resolved the connection the hum level was back to where it previously was. Check your bias circuit connections and setting. Not disputing the movement of the PT either though.

btw, haven't heard an East Coast accent like that since 79. I'm going to guess CT-MA-RI boarder region.

silverfox.

[basschops1528]

I'd like to think moving the OT is last resort after a shielding attempt and one more thing, DC filaments.
I'd feel comfortable putting in a small recto circuit because that would almost eliminate any chance for hum globally. Ideas?

[MFowler]

I don't think the OT placement has anything to do with this hum.  I've placed OT's the same way in a lot higher gain amps then this one without problems.


Grounding of preamp to one spot.  Ground wires of the pots need to be checked you had dc on the volume pot in your video.


PI & the rest of the filter cap grounds to one spot near PT.


Did you ground the speaker jacks?


Mark

[sluckey]

I'd like to think moving the OT is last resort

I know it's a pain to move it and it may not even be possible on your chassis. But, good layout is the FIRST ORDER of business when designing a hum free amp. In my early days I have totally abandoned a chassis and started all over with a new chassis and a proven good layout.

You could temporarily try this. Just totally remove the OT and set it on the bench well away from any sensitive preamp circuits. I'd put it close to the power tubes. Then use jumper wires to connect it. Does the hum problem get better?

[Paul1453]

I vote for "move the OT away from that preamp tube".

You have your out-house right next to the drinking-well. An extended poop-pickup test would just confirm the obvious: move the outhouse.

Shielding is rarely effective. The metal can on the first tube is not bad, but won't keep heavy crap out.

This analogy makes sense to me, and is easy enough to remember.   

Who wants a drink of poopy water?   

You can try to filter out the poop if you want, but why let it get in there in the 1st place?

So, now when I'm doing a new layout, when  pick up the OT I ask myself that question.

And I think long and hard about where I want to put my outhouse.      

[basschops1528]

Sluckey,

Just an electricity question. Is the hum in the OT coming from ac ripple in the B+?  I think not because its 60Hz and not 120Hz to my knowledge.

[Paul1453]

Sluckey,

Just an electricity question. Is the hum in the OT coming from ac ripple in the B+?  I think not because its 60Hz and not 120Hz to my knowledge.

I'm nowhere near as knowledgeable as Sluckey, but:
That B+ fed to the OT is bumpy DC or MMMMs as I sometimes call it, and the peaks are at 120Hz. 

Your AC heater supply is a likely source of 60Hz hum.

[basschops1528]

I will have to double check how it sounds but I'm pretty sure its 60 cycle., at least what I'm hearing

[2deaf]

The hum in the video is coming through my computer as 120Hz.  Anybody else hear it like this?

[HotBluePlates]

I wasn't sure what I was hearing. So I converted your video to an mp3, and isolated the sections where the hum is heard but you're not talking or playing.

Spectrum analyzer shows most hum heard is 120Hz and its harmonics; that stuff can only come from your B+. But there also a much lower-level 60Hz hum, which would come from your heaters, very likely the heater wiring.

See attached screenshot.

[Bluemeany]

I wasn't sure what I was hearing. So I converted your video to an mp3, and isolated the sections where the hum is heard but you're not talking or playing.

Spectrum analyzer shows most hum heard is 120Hz and its harmonics; that stuff can only come from your B+. But there also a much lower-level 60Hz hum, which would come from your heaters, very likely the heater wiring.

See attached screenshot.

I've not seen audacity used like that before HotBluePlates. That's a great way of determining the source of problem whines, whistles and hums!

[silverfox]

So if one were to use an Oscilloscope to look for the source of hum, what level of 60 hz hum on the preamp section would begin to show up in the output? If this is aj open ended question, how would you use a scope to isolate the hum source?

regards,

silverfox.

[Paul1453]

I would use the scope to look at the AC ripple in your PS.

B+ is always bumpy DC, it has the least filtering.

More forward one LC or RC stage at a time.  If that stage's E cap is working well you should see a marked decrease in the size of the bumps.

Each stage your AC ripple should get smaller and smaller.

Apparently guitar amps don't need, or really even want completely filtered and regulated DC voltages.

I guess some of those bumps and sags are needed by the tubes to give us those sounds we love.   

So your always going to have some AC ripple on your B+, B+1, B+2 voltages.

I'd say as long as you see a marked decrease of ripple in each stage as you go forward, the E-caps are doing their job.

After verifying the PS E-caps are all working as planned, I would switch to a listening device like Doug shows in his tips/tools section?

A signal tracer or other listening device is really helpful to find your hum problems.

You stick your listening probe in suspect areas, and see what you hear.

If the hum is bad enough you will definitely find hot spots of hum with your listening probe.

Once you've located your trouble spots, the fun begins.

It is not always easy to figure out why the hum is bad in that spot.

But even if you can't find and completely eliminate the problem, at worst you could try to reroute wires or add shielded wires in the hot spots.   

That's just the way I tend to approach hum problems.   

I'm sure others have their own different methods, that might even work better?   

[sluckey]

So if one were to use an Oscilloscope to look for the source of hum, what level of 60 hz hum on the preamp section would begin to show up in the output? If this is aj open ended question, how would you use a scope to isolate the hum source?

Just set your sig gen to 60Hz, connect it to the input and start measuring.

[basschops1528]

Is it possible to connect a scope to the B+ to view the AC ripple? I figure if I can get a view of what it looks like I can probe other places to find when its getting in and try the shielding
 Now that I know its most definitely 120Hz hum, I need to see where near V2 its getting in. (Again, with V1 out and V2 in there's hum, with both out or just V1 in there's no hum. V3 obviously stayed in

[Paul1453]

Is it possible to connect a scope to the B+ to view the AC ripple?

Absolutely!  That is the best way I know to accurately measure what your AC ripple on each voltage is.   

[basschops1528]

So how do I go about doing that without blowing up my scope that I couldn't possibly replace easily? aren't the inputs only rated to handle like 100v? I'm gonna assume there's a thread on this already....

[HotBluePlates]

... aren't the inputs only rated to handle like 100v? ...

Look at your scope's manual and/or specs. Most scopes would be rated to withstand 400-600v on the input, but that depends on your scope's input circuit.

And you'd probably want to use the a.c. coupling mode so you only see the ripple, not the big d.c. voltage offset...

[basschops1528]

instead of just using the 1x or 10x mode can't I use a resistor in series to do some limiting. I'm just extremely worried about taking that kind of measurement without being 110% sure I know what I'm doing is right without a tech right there.

[basschops1528]

Ground wires of the pots need to be checked you had dc on the volume pot in your video.

What would be causing DC on the pots and where can I measure it with a MM?

Did you ground the speaker jacks?

I did ground the speakers jacks, the wire runs all the way to the PT single ground. They are isolated from the chassis. Grounds for input jacks and preamp caps are in between each on the chassis bolted.   

I will take pictures

[2deaf]

instead of just using the 1x or 10x mode can't I use a resistor in series to do some limiting.

How about a 0.1uf @ 630V capacitor in series?

[pompeiisneaks]

> a piece of stainless about 1.5mm thick.
> an electromagnetic pickup with a resonant frequency at 60Hz?

I vote for "move the OT away from that preamp tube".

You have your out-house right next to the drinking-well. An extended poop-pickup test would just confirm the obvious: move the outhouse.

Shielding is rarely effective. The metal can on the first tube is not bad, but won't keep heavy crap out.

PRR your genius makes me laugh and learn at the same time... This sounds like the wisest yoda saying I've heard in a while ;) 

[basschops1528]

I've found something interesting.

V2 is the one having the interaction, I just know it is. I just looked under the board and you guys are gonna love this. Tell me what you make of this...

The brown and Red wires from the OT going to the Power Amp section run under the board just below the preamp tubes and to the left of all the tube sockets.

It just so happens that I didn't secure it well enough and it's at least 1 cm away from the board with its highest crest near the V2 caps and resistors.

The presence return from  the output jack runs about parallel to it

The presence pot makes scratchy noises when you turn it and one final scratch when opened wide.

Now MFowler said I had DC on my volume pot. I did notice that Vol1 does change the sound a bit (but not the hum as far as I can tell), but wouldn't you say the presence pot has DC on it?

What if the AC Ripple from the OT Primaries, carrying the B+ from the CT is inducing that hum into the presence return, which is being fed back through and it's "DC" enough to make scratchy sounds and produce hum, plus any fields coming from those Primaries into the components above the board?

I can't  explain the DC on the Volume though. My grounds seem solid. Either bolted, soldered, or both.

[Paul1453]

... aren't the inputs only rated to handle like 100v? ...

Look at your scope's manual and/or specs. Most scopes would be rated to withstand 400-600v on the input, but that depends on your scope's input circuit.

And you'd probably want to use the a.c. coupling mode so you only see the ripple, not the big d.c. voltage offset...

Yes you do need to know what your scope is capable of measuring.
We cannot tell you what that is.

Now it has been over 25 years since I did this on a daily basis, but this it how I remember it going.
OK, so we are using a scope capable of measuring at least 500VDC.
So we have the scope set to measure DC at our largest interval possible.
We connect the probe to the B+ and find our trace.  Then we center that trace on the middle line.
Then we switch the scope to measure AC.  Since our measurement interval is still set to the highest possible, we don't see any AC yet.
Still just a flat line at the center of our scope.  Then we start dialing the measurement interval down until we start to see waves instead of a flat line.
We may need to adjust our sync at this point to start to see the waves instead of a fuzzy fat line.
But we keep dialing our interval down until we see the waves or the fuzzy fat line.

So lets say for this example we expect to see 4VAC ripple on our just rectified DC.
When we get down to the 1V per division setting, we should have an AC wave that measures 2 divisions above and below the middle line of our scope.
The gear I worked on would usually give us a spec for acceptable AC ripple on each voltage.
If we expected to see 4VAC ripple here, but when we dialed down to 10V per division we got the wave measuring 2 above and 2 below the line, we have a problem.
Instead of 4V of ripple we are measuring 40V.  Our first E-cap is not doing it's job.

Instead lets say we dial down to 1V per division and we see our wave measures 2 above and 2 below or less.  So we are good here.
Now if we move our probe to the next voltage test point, we may lose sight of our trace if the next DC voltage is much lower than our first.
Our DC offset is set to put the trace in the middle for our 1st voltage.  Our trace might have dropped off the bottom of our screen.
So we could just adjust our offset lower to put the trace back in the middle of the screen again. 
But if we can seem to find the trace now, we can always start over on DC measurement largest division setting to find our trace and put it back in the middle of the screen again.

So now we expect our AC ripple to be lets say 2V at our next voltage.
When we get dialed down to 1V per division and we see our AC wave 1 above and 1 below the middle line, we are good.
If we still have 2 above and 2 below the line we know the second E-cap is not doing it's job.

So our 3rd DC voltage we expect to have 1V or less AC ripple.
We follow the same procedure, but this time we should be able to dial down to .5V division and see our AC wave 1 above and 1 below the center or less.

I hope you followed my example.  Like I said it has been a long time since I last did this.
When you do it on a daily basis, it becomes like 2nd nature and you really don't need to think about the dials and knobs you turn to make it happen.
It makes it harder to explain to someone else then, but very easy to just do it yourself.   

[pompeiisneaks]

I wasn't sure what I was hearing. So I converted your video to an mp3, and isolated the sections where the hum is heard but you're not talking or playing.

Spectrum analyzer shows most hum heard is 120Hz and its harmonics; that stuff can only come from your B+. But there also a much lower-level 60Hz hum, which would come from your heaters, very likely the heater wiring.

See attached screenshot.

HBP, I'm glad you showed me this, I use audacity all the time and didn't realize you could do this!  So cool.  I'm going to remember this from now on, I pulled up a previous sample of my ac30 and can see it has a little more 60 than 120 but neither is horrible, (sub -50db, but I def would love to play with this more). 

As for the comments about scopes, please someone correct me if I'm wrong, but part of what the scope can handle depends on the probe itself.  If it is a 1x probe, then the scope can handle what the input specifies.  the 10X probes can then handle more voltage, but that's defined by the probe itself, and for my use I've bought a couple 100X probes that can handle something like 2000V so I should be fine.  I think most 10X (read the manual) can go up to about 500 or 600 which should then be fine, but you definitely want to be wary of input spikes due to power on etc,  This is why I just went with the 100X probes.   Here's an example probe I bought: http://smile.amazon.com/gp/product/B0030L0X58?psc=1&redirect=true&ref_=oh_aui_search_detailpage

I wouldn't recommend buying that probe, though, the vendor is pulling a fast one on you.  The picture is a hantek. The actual ones I got were black and much lower quality looking.  I think I'd buy something that costs a bit more in the future, but these do work.

~Phil

[HotBluePlates]

I've found something interesting.

V2 is the one having the interaction ... The brown and Red wires from the OT going to the Power Amp section run under the board just below the preamp tubes ... with its highest crest near the V2 caps and resistors. ...

I'm glad you're confirming this for your own insight. This is what Sluckey was referring to way back in Reply #1 (as well as Eleventeen ion Reply #9 and PRR in Reply #11):

... I would move the OT away from that preamp tube. Put it near the output tubes. ...

The presence pot makes scratchy noises when you turn it ... MFowler said I had DC on my volume pot. ... wouldn't you say the presence pot has DC on it?

What if ...

Sure your Presence pot has d.c. on it, as does every Marshall with this same Presence circuit (as did my old '73 50w head). It's the nature of the circuit because d.c. volts are across the Presence pot due to it being between the inverter tail resistor and ground. This is not a flaw for you to fix, but normal for this circuit.

I highly recommend you solve one issue at a time, and set aside the volume control issue for now. Moving the OT is going to be your biggest effort, which you've already figured out needs to happen.

After you've got that relocated and the hum issue resolved, then you can look back at why the Volume pot is scratchy (which is going to be a d.c. issue, but may be indirectly resolved when you unsolder/resolder things to move the OT).

[HotBluePlates]

... What if the AC Ripple from the OT Primaries, carrying the B+ from the CT is inducing that hum into the presence return, which is being fed back through and it's "DC" enough to make scratchy sounds ...

Direct current cannot couple through the air, only a.c. can. This is why a capacitor blocks d.c. but passes a.c.

Just wanted to touch base on that so you would know where to look when chasing scratchiness and d.c. issues. They can only come through a direct wired connection to the problem area.

[PRR]

> Direct current cannot couple through the air

Stand next to a large Van De Graaff machine and say that.

But agree that we never get *near* such conditions, and DC-"coupling" through-the-air is a total non-issue in g-amps.

(But recall that air is a lousy mechanical support, we always have poly or bakelite or other supports, and cheap/old/damp "insulating" board can leak enuff DC to upset delicate inputs.)

[MFowler]

I said your volume pot was scratchy dc on it but that was the presence pot I was mistaken.


The presence cap needs to be in front of the presence pot pin 3 to block the dc that causes scratchiness rather then on pin 1 to ground.


I would also move some wires around with a chop stick to see if that helps.


Also, I had a hum problem once that was caused by the room I was building my amp in.  Try it in another room, watch out for fluorescent lights and those conical light bulbs they produce nasty raspy hum in the AC line.  (A spiral-shaped compact fluorescent lamp (light bulb).


Mark

[HotBluePlates]

> Direct current cannot couple through the air

Stand next to a large Van De Graaff machine and say that. ...

Yeah, I was wrong on that one. Electrostatics... And of course, d.c. pulls electrons through vacuum to a tube plate.

I agree with your restatement that d.c. coupling through air in guitar amps isn't a problem, and specifically wouldn't lead to a scratchy pot.

[basschops1528]

The long lead OT test was negative. It didn't matter whether I had the OT on a work bench ~3 feet away, furthest from it's original mounting point in the same area, or exactly where it was. The hum is the same either way. It does however seem to be a little quieter overall, but volume doesn't change with either position. I even tried turning the lights off. The only thing somewhat nearby (~5 feet) was the breaker panel, and I waited for the furnace blower motor to be off each time. I may just be able to keep the OT where it is. I think the way it was wired was synergistically adding to another lead dress issue in the system and making the hum worse ( I read something like this in another thread). I will take some pictures so you guys can see where the wires are draped atm, and attached from underneath.

[basschops1528]

So here are some pictures I've taken of the OT; center tap and primaries, and the secondaries as well. I also included some pics of the V2 socket so you can see how I wired it and of the circuit components associated with it.

[basschops1528]

More

[basschops1528]

and more

[basschops1528]

Bump

[Willabe]

What are all the long wires coming out of the amp?

[basschops1528]

Willabe,
As sluckey suggested I removed the OT and soldered long wires in so I could move the OT away from the amp. Turned out that no matter where I put the OT the hum is still there, even right next to the preamps, literally closer than when it was mounted.

[Willabe]

He didn't mean 3' to 4' long wires. Just long enough to move the OT away from the pre amp.