I can't for the life of me get rid of this hum.

[liddokun]

Hey folks,

I just finished my first scratch built amp, an ac15. 

Everything works fine after I fired it up, but there's this persistent hum.  I can't pinpoint the source, though it is present when there's no guitar hooked up. 

I'm lead to believe that it's ground related; I've tried to do as much reading as I could on this and went back to tweak my grounding scheme a number of times. I'm grounding using buss grounding.  Each section of the amp goes to it's respective filter capacitor, then each capacitor is connected to the buss in this order (from furthest away from chassis ground) power amp, PI, vibrato/trem channel, normal channel.  The buss is grounded to chassis next to the input jacks, which are also isolated from chassis. 

OT common is sent to output jack sleeve, and then a wire from sleeve is sent back to the bus. 

Could it possibly also be my lead dress/heaters are not well done?  I am not using sluckey's layout, I built this in the more traditional AC15 layout with the poweramp in a seperate chassis. 

Here are some photos of my heaters, could this be the issue?


Edit:  The next thing I'm planning to do is add a humdinger pot to try and balance heaters. 

[PRR]

Welcome.

What knobs affect the hum how?

Show us (link-to) the AC15 plan you are working from. (There are just enough variations to sow confusion.)

[liddokun]

Here's a link to the layout I worked off of.

http://members.ziggo.nl/jheijer/bestukin1.htm

The hum is present when no signal is plugged in;  none of the knobs effect the hum.  Cranking the amp up adds noise, but the added noise sounds like single coil noise to me. 

[Leevi]

Some basic things first:


Is the heater circuit balanced with resistors or center tap or grounded?


Are you using choke?


By changing/combining the grounding points may help.
I have built successfully several AC15 by using bus and only one grounding pot


/Leevi

[liddokun]

Hey Leevi,

The heaters have no center tap; I'm running 2 220R resistors to the cathode of the first el84. 

I was contemplating installing a 250R potentiometer to better balance the heaters.

I am using a choke. 

What changes to grounding do you suggest?

As it is now, I've got the filter caps hitting the buss as follows:

Poweramp----Phase Inverter----Channel 1----Channel 2------chassis ground (near isolated input jacks)

[John]

Something easy to try: change your first filter cap ground (coming off PT) to a bolt that holds down the PT.

[Leevi]

What changes to grounding do you suggest?As it is now, I've got the filter caps hitting the buss as follows:Poweramp----Phase Inverter----Channel 1----Channel 2------chassis ground (near isolated input jacks)

I would try the following combinations:


1. One common preamp point and one common poweramp (including PI) point
2. One common point to the whole amp


It can also help if you add an extra filter cap and a low ohm power resistor before the first filter stage (B+ stage)


Connect an AC volt meter to the output and check the hum-voltage after each change.


/Leevi

[sluckey]

The 120hz ripple on the first two filter caps is many, many times greater than the ripple seen on any of the other filter caps. It is very important that the 'dirty' current for these two caps does not travel the same path that the sensitive preamp ground currents must follow.

Both of those power amp filters (one on each side of the choke) need to connect to chassis near the power transformer. The HT center tap needs to connect to the chassis at the same point as the power amp filters. Power tubes cathodes, PI grounds and filter need to connect to this same chassis ground point. OT secondary and speaker jacks connect to this same chassis point. Call this the power ground.

All input jacks, preamp grounds, and filter caps need to connect to a buss that connects to chassis near the input jacks. Call this the preamp ground.

Best not to mix the power grounds and preamp grounds. Even though the PI is usually considered part of the power amp, it's sometimes more logical/convenient to put the PI grounds to the preamp ground. That's OK. But by all means, keep the PT center tap, first two filter caps, and power tube cathodes grounded at the power amp ground point near the PT and far away from the preamp ground point.

[jjasilli]

there's this persistent hum.  I can't pinpoint the source

Always use flowcharts & checklists to trackdown isues!!!  Tubenit has posted some good ones on this Forum.


Anyway: 
Determine the frequency of the hum. It's either 60Hz or 120Hz.  120 is pretty close to a B note on the guitar's A string.  60Hz indicates heater hum.  120 indicates a Power Supply issue.  Hum can also arise in a Fixed Bias circuit.  In the last case, hum will not vary with the Vol control.  In fact, if hum does not vary with Vol, then the hum arises after the Vol pot.

[shooter]

don't "overlook" new tubes being bad

[liddokun]

Thanks for the suggestions everyone, I'm going to redo the grounding to separate power amp from preamp.

I'm fairly certain its 60Hz, it's lower frequency sounding.

shooter, I'll try switching out tubes with known good tubes as well to see if this helps. 

Ordering the humdinger pot to balance heaters as well.

[bnwitt]

Youtube 60hz hum file.  You may have to turn up your computer audio to hear it.

http://www.youtube.com/watch?v=SVZ2P0KsLic

Youtube 120hz sine wave

https://www.youtube.com/watch?v=pC4Uzt0qm2E

[trobbins]

Can you put a microphone near a test speaker and record the hum, and then use PC software spectrum analyser to see the hum frequencies?  That's a pretty crude test.  Can even do that with an android cell phone running free spectrum analyser software.  If you're keen on amps then many of us set up a probe in to a PC soundcard to get much better 'performance'.

[liddokun]

I'll give that a shot.  Fairly certain it's 60hz though.

A suggestion from a local tech to pinpoint where the hum is coming from by pulling the phase inverter tube. 

So with no PI tube the hum is still present. 

[John]

Pull your power tubes. If hum still present, then your OT is picking up bad vibes from your PT. I had this happen to me and it drove me nuts till I figured it out.

[liddokun]

So the hum goes away when the power tubes are pulled.  So I guess my issue is somewhere in the output/power side of things. 

EDIT: At the suggestion of trobbins, I recorded the hum and ran it through a spectrum analyzer.  It basically told me that I am a fool and what I was convinced was 60Hz hum is actually appearing to be 120Hz hum. 

So what I know so far:

Hum is present with no guitar plugged in, hum is present when phase inverter tube pulled, hum is not present when output tubes are pulled.  It is in fact 120Hz hum. 

[trobbins]

There seem to be a few AC15 schematics, so perhaps if you can link to the circuit you are using?

Some typical tests to do:

Add cathode current sense resistors for each EL84 from the top of the common cathode bias node, and confirm that they have balanced idle current.

Add temporary hardwire links from EL84 grid to 0V node of common cathode bias.

Add a temporary RC to filter your screen supply.

[sluckey]

hum is present when phase inverter tube pulled, hum is not present when output tubes are pulled.

That narrows it down quite a bit. Gotta be in the output circuit or power supply circuit. Show us some hi-rez pics of the power amp and power supply chassis wiring. Need to see the ground wiring for power transformer CT and filter caps and output tube grid and cathode grounding.

Try this... Use a couple gator clip test leads to ground both 1.5K resistors that connect to the output tube pin 2. You can ground either end but do so for both resistors. Hum go away?

[liddokun]

I tried grounding both 1.5k resistors (on the non pin 2 side of each resistor).  The hum remains.

Here is a photo of my output/power chassis wiring.

A local tech had suggested separating my cathode ground from my first filter cap ground; right now the cathode is being grounded to the first filter cap then to the chassis.  Would this help?

[trobbins]

You haven't taken the HV CT to the first filter capacitor negative terminal (the cap seems to be a combo and only have one negative terminal).  Highly likely that is your major mistake.

Where is the OT B+ lead?  It doesn't seem to be anywhere near the EL84 anode wires, or directly off the second filter cap positive terminal.

Perhaps re-route the 'blue' EL84 anode wire well away from the other EL84 valve base (especially pin 2).

How do you know the EL84's are matched for idle current ?

The screen stoppers are right next to the grid wiring, and the B+ also (at the other end of the screen stoppers).   Perhaps route the grid wires from the valve base along the chassis so they are as far away from screen wiring as possible.

[trobbins]

Also its wise to just have one connection to chassis from all your secondary side circuitry.

[pdf64]

Yes, well spotted, it looks like the red/yellow wire will be carrying significant ripple current.
So the entire HT system will have that ripple current superimposed on its 0V return.
Even though the resistance of such a short piece of wire is low, the current peaks (from charging up the reservoir cap every half cycle) will be high.

[liddokun]

@trobbins:

The HV CT was initially going to the first filter capacitor negative terminal in my original wiring; at sluckey's suggestion a few posts above, I collected all the power and output grounds at one point instead and ran them to the chassis near the PT.    Actually, my initial grounding was to have everything on the output/power side run to the first filter capacitor negative terminal, then that to chassis. 

OT B+ is the run along the top of the chassis (you can kind of see the red wire poking out), and to the two resistors on the right side of the turret board.  Then that's run to the standby switch, and the switch returns a green lead to the filter capacitor.  I'm using this schematic, which places the standby switch after the choke, as oppose to before it in Sluckey's schematic. 
http://members.ziggo.nl/jheijer/voeding.htm  Not sure if this makes any difference functionally.

I'll try rerouting the anode wire away. 

I don't know the EL84's are matched for idle current; I didn't buy a matched pair, but I'm thinking I should buy a matched pair now. 

As for routing the grid wires away from the valve base, I'm a little new to this, would you mean the wires coming from pins 2 and 3 of the output tubes?

[shooter]

I'm a little new to this,

these are real handy for quick reference;

[pdf64]

The idle dc in either leg of the OT primary need to be reasonably similar, in order for the HT ripple there to be 'common mode' and so cancel out (due to the 2 primary halves being in opposite polarity); if there's much discrepancy, a good degree of cancellation can't take place and ripple will appear.
Hence with shared bias arrangements, it's best for tubes to be matched, ie for them to be fairly similar plate currents for a given control grid voltage.
If an independent cathode bias resistor for each tube is used, tubes tend to match themselves, ie the self regulation of cathode bias tends to idle tubes at fairly similar plate currents.
As per the data sheet, 270 ohms per tube, each with its own bypass cap.

[liddokun]

updates:

I went back and made a few changes based on the suggestions.

1. HV CT run directly to the negative terminal of the first filter cap.  Negative terminal of filter cap then run to a point on the chassis near the PT.  PI ground, sleeve of output jack, cathode RC ground points also to this point on chassis.

2. I've removed the standby switch completely, OT B+ goes directly to 2nd filter cap positive terminal.

3.  Rerouted the blue anode wire away from the other el84 valve base. 

Plate voltages are 340V.

Hum is still present, unfortunately.

[pdf64]

Until you get matched tubes (or move to independent bias), how about assessing the current in each leg of the primary, eg measure the resistance of each, then the voltage drop across each?

Your HT seems rather high?

[liddokun]

Any suggestions for a bypass cap value if I were to go to 270R individual cathode resistors?

On Sluckey's schematic the reference plate voltage is 331v, I'll go through and measure everything else though and see what the actual plate dissipation is.

I sent my tube guy a text to see if he's got any matched pairs in is current inventory. 

Could this be a possible cause of my hum as well?

[pompeiisneaks]

Any suggestions for a bypass cap value if I were to go to 270R individual cathode resistors?

On Sluckey's schematic the reference plate voltage is 331v, I'll go through and measure everything else though and see what the actual plate dissipation is.

I sent my tube guy a text to see if he's got any matched pairs in is current inventory. 

Could this be a possible cause of my hum as well?

Bypass caps on split cathodes can be identical, they're just creating a specific filter based upon the uF value.  Unless you want different frequency response per tube. 

I was reading this and didn't recall seeing if you'd separated the preamp grounds from the power grounds like sluckey mentioned?  I had a decent hum with my AC30 I first built due to not separating grounds between preamp and power sides.  It literally disappeared once I did so.

~Phil

[liddokun]

Yes, I've separated the preamp and poweramp grounds. 

I think we've determined that the hum is coming from somewhere in the output/power section because the hum persists when the PI tube is pulled but goes away when the output tubes are pulled, though the hum comes back when only one output tube is pulled. 

There was a query about whether or not my output tubes were matched, or if they were even known to be good "e.g. tested in a different amp with no issue".  The answer is no, they are NOS soviet tubes of a Russian site.  I texted my tube guy and he is going to find a matched, tested pair for me. 


For bypass caps, what's a typical uf value used in this situation?  Say for a pair of 270R cathode resistors?   My plate voltages seem to be sitting a little on the high side, might be good fiddle with the cathode bias resistors a bit too.

[shooter]

what's a typical uf value

It's a to flavor kinda thing, Bass gets tighter the higher you go.  I use 47uf as typical, but have used from 22uF to 220uF

[liddokun]

So I popped a matched tested pair of output tubes in the amp.  Hum is still kicking.   

[trobbins]

How did you know they were matched in your actual amp - a measurement of some kind when in idle state?

As indicated previously, this is very easily done with a separate cathode bias resistor, or a separate sense resistor going to each cathode.


Have you temporarily added an extra RC filter to the screen supply, to check that influence?

[liddokun]

My tube guy measured said tubes on his tube tester. 

Separated the cathode bias resistor,  235R resistor each to ground from each cathode.

I'm getting 331v plate voltage on both output tubes, and 10V at the cathode for each output tube.  Do these measurements look about right?

@trobbins, where would I place the RC filter and which values are typical?

[Leevi]

Do these measurements look about right?

Yes, they are fine


/Leevi

[trobbins]

Each EL84 has a screen stopper, but the stoppers are directly connected to 315V B+. The ripple on the B+ can become noticeable if directly connected to the screens.  Often the screen supply is further RC filtered, often as a way to drop the screen voltage.  I'm not sure what you've got lying around, but an example would be a 1kohm and a 50uF 400V filter cap to supply the screen stoppers.  That would give a 3Hz corner frequency, so should heavily filter any hum.


Faultfinding is often about steps that include or exclude a possible issue (eg. if the output stage valves were grossly mismatched).

[liddokun]

I do happen to have a few 47uF 450V caps lying around, I'll give it a shot and see how it goes. 

[liddokun]

Updates:

I tried adding an extra RC filter supplying the screen resistors.  I used a 7 watt resistor, it started to give off smoke so I cut the power pretty quickly.  But that did not help the hum.

What DID help the hum; I went back and switched the output jack to an isolated cliff style jack.  This cut a lot of the hum out, so I guess I can be happy about this.  I have my output jack and tremolo on a separate steel back plate.  I assumed this was enough to "isolate" the output jack and used a metal switchcraft jack. I guess I was getting some kind of a ground loop here.  I need to tinker with this some more tomorrow. 

[John]

Updates:

I tried adding an extra RC filter supplying the screen resistors. I used a 7 watt resistor, it started to give off smoke so I cut the power pretty quickly.  But that did not help the hum.

What DID help the hum; I went back and switched the output jack to an isolated cliff style jack.  This cut a lot of the hum out, so I guess I can be happy about this.  I have my output jack and tremolo on a separate steel back plate.  I assumed this was enough to "isolate" the output jack and used a metal switchcraft jack. I guess I was getting some kind of a ground loop here.  I need to tinker with this some more tomorrow.

This isn't really on topic about your hum issue, but that 7W should have been more than big enough.

[liddokun]

hum is present when phase inverter tube pulled, hum is not present when output tubes are pulled.

That narrows it down quite a bit. Gotta be in the output circuit or power supply circuit. Show us some hi-rez pics of the power amp and power supply chassis wiring. Need to see the ground wiring for power transformer CT and filter caps and output tube grid and cathode grounding.

Try this... Use a couple gator clip test leads to ground both 1.5K resistors that connect to the output tube pin 2. You can ground either end but do so for both resistors. Hum go away?

Hi sluckey,

If the hum goes away when both these resistors are grounded, what would be my next steps in figuring my issue out? 

[pompeiisneaks]

hum is present when phase inverter tube pulled, hum is not present when output tubes are pulled.

That narrows it down quite a bit. Gotta be in the output circuit or power supply circuit. Show us some hi-rez pics of the power amp and power supply chassis wiring. Need to see the ground wiring for power transformer CT and filter caps and output tube grid and cathode grounding.

Try this... Use a couple gator clip test leads to ground both 1.5K resistors that connect to the output tube pin 2. You can ground either end but do so for both resistors. Hum go away?

Hi sluckey,

If the hum goes away when both these resistors are grounded, what would be my next steps in figuring my issue out?

That to me says he was trying to rule out power tubes or power supply, since you just grounded the signal at the input and the noise went away, it's likely in the power tubes themselves and not the power supply circuit.  If I'm gathering correctly the point of his test.  He's  going to likey chime in as well, though :) so I'll wait to see if I'm right heh


~Phil
 

[liddokun]

That to me says he was trying to rule out power tubes or power supply, since you just grounded the signal at the input and the noise went away, it's likely in the power tubes themselves and not the power supply circuit.  If I'm gathering correctly the point of his test.  He's  going to likey chime in as well, though :) so I'll wait to see if I'm right heh


~Phil

That makes a lot of sense.