How do I deal with a failed electronic project?

[Planobilly]

Hi,
My last amp design/build has failed to operate as planned. There are several perplexing issues I do not know the cause of. I started the amp up and it will play and has all the voltages I expected within a few volts.

Issues
1. Serious amount of 60hz hum.
2. Channel two gain and volume control interacting in a very non standard way.
3. Every component in the preamp stage is microphonic. I have never seen a amp do that. What I mean by that is when I tap on a component I hear the tap sound in the speaker.

What I had in mind with the design..
I am trying to combine a channel from an old Fender Bassman with a channel from a modern high gain Fender EVH 5150. The amp is class AB push pull 100 watt.

As I have invested a large amount of time, effort, and money in this amp, I am not very happy with the prospects of accepting failure as the outcome.

I assume I can find and fix the hum issue. Perhaps that will not be easy but...
I assume I can fix the gain/volume control issue...
I have absolutely no idea what could cause a seven stage preamp section to display a microphonic condition. Obviously there is not enough information in this post for anyone to comment as to the exact causes of the above faults. I can post all the schematics and give a detailed assessment of the faults with oscope screen shots where needed.

My question at the moment is where do I start? I assume I should try to fix the hum issue first.

Is that the best place to start?

Thanks,

Billy

 

[Tom.G]

How do I deal with a failed electronic project

TROUBLESHOOT.

"Serious amount of 60hz hum."
I would expect 120Hz hum, from the power supply. 60Hz indicates either an open input, wire routing, or maybe a ground loop.

should try to fix the hum issue first.

Is that the best place to start?

Yup! That's probably the easiest one to fix. Start pulling tubes sequentially from the input stage to the output, while paying attention to the hum level and waveform.

The microphonics problem is likely either the gain is too high or you got a bad batch of tubes. Or it may be a corroded tube socket or connector, or even a solder joint that wasn't soldered.

A readable schematic would be useful.

p.s. All of the problems *could* be a missing ground connection somewhere in the build.

 

[Planobilly]

Thanks Tom,
I will try to clean up the schematics with the changes I made tomorrow. No one else but me could read them at this point.
The hum is reduced as I pull preamp tubes. It increases with gain and volume.
The tubes are known good and tested in another amp and I have moved the tube sockets around enough to be pretty certain that is not the issue. I have disconnected things looking for issues, so I need to get back to "normal" and check every solder joint and every component connection.

This is the first amp I have built that had any real issues. It is by far the most complex also. I think the design will work but...
The amp will actually play and all the voltages on the tubes are pretty much dead on and no smoke is coming out...lol...so there is some good news!

Thanks,

Billy

 

[davenn]

My question at the moment is where do I start? I assume I should try to fix the hum issue first.

Is that the best place to start?

yeah, definitely, you don't know what effects that may be having on the other issues

1. Serious amount of 60hz hum.

is it really 60Hz or is it 120Hz as Tom suggested ?
Scope the ripple on your PSU see how bad it is

disconnect the power amp stages and scope and listen to the pre-amp output is there hum ?

A readable schematic would be useful.

agree with Tom ... show us what you are up to Dave

 

[jim hardy]

Troubleshooting. try to draw a circle around the problem and then tighten the noose..

on your hum... to rule out one area, heaters a couple thoughts to ponder

7199 is an RCA tube designed for early preamp stages to minimize hum from heater-cathode capacitance..

i'd try lifting the preamp heater supply leads and powering them temporarily from a 6 volt battery.

i've seen heaters grounded via a ~hundred ohm pot across heater supply with wiper grounded for adjustable hum reduction.

remember the old trick of using preamp heaters as output pentode cathode resistors , that gives you more nearly DC heater power and prevents applying signal to not-yet warmed up drivers.

old jim

 

[Planobilly]

Here is the schematic showing the PA and the HT part of the power supply. The B+ loaded voltage is 480 VDC with 5.19 VAC ripple. There is very little ripple past the B+. This ripple is 120Hz.

Here is the Preamp. The channel switching is not installed yet and the effects loop is not connected to the jacks.

Billy

 

[Tom.G]

I see that the FS Jack Ground is connected to Chassis Ground, not to the Power Supply common. If this isn't a problem now, it could be when the rest of it is straightened out.

The channel switching is not installed yet

Which preamp channel is being used?

The hum is reduced as I pull preamp tubes. It increases with gain and volume.

If that means the hum progressively decreases as more preamp tubes are pulled, it indicates the hum is introduced into ALL the preamp stages.

Next thing to look at:
1) Using the scope, verify the frequency and waveform of hum at the output. (photo would help)

Is this the build you presented in an earlier thread looking for comments, the one with DC Heaters from a poorly filtered full wave rectifier?

i'd try lifting the preamp heater supply leads and powering them temporarily from a 6 volt battery.

Certainly worth the effort! Good one Jim.

 

[jim hardy]

C21 ?

 

[jim hardy]

another idea to poke with

imagine a tube big enough you could walk into. Like at some radio stations..

The grid would be like a barbed wire fence that you could shake.

Tubes are microphonic because moving the grid wires modulates the electric field inside.

Take an old fashioned wood pencil that's sharp and gently tap tubes and components. I predict you'll find a particularly sensitive spot probably on the glass envelope of an individual tube.

Is this the amp you pictured with parts so neatly mounted on turret boards ? It's going to feel great when you get it straightened out.

In my projects Mother Nature always makes me struggle. She is one coy mistress. Persist.

 

[Planobilly]

I solved the major hum issue.
The preamp microphonic issue went away with the repair of the bias circuit
The bias circuit was incorrectly wired. Two of us were working on the amp. I did not check everything Dave soldered up on the board. The channel one circuit is not working correctly. It is really distorted. It is a simple circuit so I should be able to figure it out.

Actually I have been at this too long and need a break to clear my head. There is also no guarantee at this point that this design will ever work properly.

I have in mind to power the heaters with a six volt battery to see what the results will be. I may have something messed up in the bridge that feeds the preamp tube heaters.

Thanks Jim for the reminder.

Cheers,

Billy

 

[Planobilly]

Hi Tom,

I have made some changes to the original schematics I posted. I will post some photos later today or tonight. I want to do the six volt battery test next. I have to go buy a battery.

Thanks,

Billy

 

[tech99]

Pre amp heaters as OP stage cathode resistors. Thank you, after all these years I never thought of that!

 

[Planobilly]

What do I need to add to filter this bridge properly?

thanks,

Billy

 

[Tom.G]

What do I need to add to filter this bridge properly?

What is the load current?
Any idea how much hum (ripple) is allowed? (Without doing the math, about 2.7V_PK-PK ripple would be around 6.3V_RMS.

BTW, I get 7.5V_PK as the output. (Two diode drops from peak AC input.)

 

[Planobilly]

Hi Tom,
I have the bridge built as above. Unloaded it measures 8.31 VDC RMS unloaded. The 6.3 VAC tap on the transformer is also powering the four 6l6gc output tubes (AC) which I did not indicate on the schematic above. With the seven 12AX7 preamp tubes connected the voltage is 6.3 VDC. I have around 500 mv of ripple. I think something is not correct in the bridge because when I set the scope to DC, I should see pulsating DC and not a sine wave. I don't understand how a AC signal could be coming through the bridge. For sure I have 6.3 VDC coming from the bridge loaded, measured with both the scope and a DMM. I also have a 120hz AC @ .48VAC

In some testing I did, I hooked up a regulated power supply set at 6 VDC (connected to the 7 preamp tubes) and the current was around 1.8 amps as I remember. I was a bit surprised as I thought it should be drawing more current than that. Actually I just looked at the datasheet again and it shows .3 per tube so the 1.8mv total may be a real number. The heaters can be run AC or DC. The whole idea of the DC heaters was to reduce hum.

In some things I have read somewhere between 4700uf and 10000uf is "normal" filtering.

Something is messed up here!

Billy

 

[Baluncore]

As much as I don't want too I need to take the main circuit board out and record the jumpers on the back side and make an as built drawing.

Your circuit board has many connections on the tubes side, less on the knobs side. If you ran the wires on the knob side under the board, then round and over to the appropriate point, you could hinge the board along the tube edge without disconnecting anything. That would make service easier.

 

[Nidum]

The hum may be being caused or made worse by the style of wiring . All those loops and crossing wires could be picking up mains frequency and other stray signal inputs and feeding them back into the system .

 

[Planobilly]

After a lot of investigation, I have come to the conclusion that the whole power supply needs to be redesigned. I am going back to AC heaters for all the tubes. As far as the filament supply, DC on filaments are most likely never worth it. The unbalanced DC hum finds it way into the signal more than a balanced AC filament supply does.
I see that high gain amps like Mesa Boogie have gone back to AC filaments after trying DC. In some cases they may run DC to the first preamp tube only. Also, all these high gain designs are built on PCBs and no one that I am aware of is attempting to build these sort of designs on a turret board. Only crazy people like me...lol

I have the amp running on both channels at the moment so I think the basic circuit design will work with some additional modification. The channel two volume control works in a very strange way which may be some error I made or perhaps is just normal for the design...not sure.

Complex amp...pushing me to and past my limited skill sets!...lol

Cheers,

Billy

 

[dlgoff]

The channel two volume control works in a very strange way ...

How so? Thinking:

Logarithmic taper potentiometers are often used in connection with audio amplifiers, as human perception of audio volume is logarithmic.

https://en.wikipedia.org/wiki/Potentiometer#Resistance.E2.80.93position_relationship:_.22taper.22

 

[Planobilly]

Hi Don,
When the volume control is set to the zero position the sound of the amp is extremely scratchy and has a very trebly sound. As the volume is increased the sound becomes normal and when increased more the bass sounds become predominate. The volume control has little effect on volume and the volume is a function of the gain control (P4 220K Pot). There is major interaction with the gain control and the treble control. Increase treble and the volume goes up. I see this characteristic with gain and treble on other high gain amps I have played. All of this may be just the "normal" way the circuit works.

If I understand the schematic the audio signal leaves V4B via the cathode through a 43k resistor into the tone stack. The gain control at P4 increases the the gain as it is turned clockwise arriving at the treble control which controls both gain and frequency. It then hits the volume control and exits the wiper on to grid of V6a. I do not clearly understand how the volume control changes the frequency of the signal other than as it is turned up V6 is provides more voltage to drive the output tubes.

I have had so many other issues to solve that I have not really given this much thought. I do know from experience that high gain amps like Mesa Boogie have controls that have a serious amount of interaction and are very difficult to get set properly. This may be the case here also.

I am considering how to redesign the whole power supply at the moment. Until I get that right, nothing else really matters much. I think that the new HT design may require a bit of complexity using zener diodes to clamp the voltage past B+ going to the grids.

Cheers,

Billy

 

[Planobilly]

Here is a possible design for the new HT part of the power supply. See any issues?

Thanks,

Billy

EDIT: there should be a couple of 220K resistors in parallel with the second set of 150 uf caps.

 

[Tom.G]

Here is a possible design for the new HT part of the power supply. See any issues?

Yup. When you figure out where to connect the Anodes of the right-most 1N4007s all the magic smoke will come out of the Zeners and maybe out of the 1N4007s.

Below was being written when the latest schematic was posted.

Inspecting the build photos again though, I find some of the suggestions may already exist. Can't tell for sure.

You keep referring to Channel One and Channel Two but the schematics don't show any such labels. Which is which?

The Power Supply schematic shows the CT of the power transformer high voltage winding thru a fuse to the bridge output. This should not be! I hope it is only a drawing error and not "as built."

The schematic also shows the 6L6 Cathodes to be floating.

I am going back to AC heaters for all the tubes.

The scope traces look as expected for the DC heater circuit. You may want to put a 100 Ohm 1/2W resistor from both the Plus and Minus of the bridge to HT minus (B-), that should reduce any hum from the heaters. Considering the scope traces, the cap values look OK.

I have come to the conclusion that the whole power supply needs to be redesigned.

I don't think that's needed; although without seeing scope traces of the various B+ voltages under load I can't say for sure.

I see the input to V1A and V2A is DC coupled though; not a good idea for this usage. Put a series 0.2uF after the input jack, this will give low freq. rolloff around 60Hz.

Another thing that may help is connecting the B- to chassis. That connection should be at the C38 - C40 junction, where the Minus bridge output connects.

 

[Planobilly]

Hi Tom,

Channel one is the low gain channel at the top of the schematic for the preamp. Channel two is the high channel at the bottom.

The 1n4007 right after the bridge is drawn in the wrong place and I will correct the schematic as the anode need to connect to the 220K resistors I left out. Sorry, not enough coffee I guess. (the last schematic I posted a few min ago)

As for the HT circuit, the center tap is connected to ground through a fuse. The two other leads are connected to the bridge. There have been so many changes that the original schematics are not as built.

Again, everything is more or less working and the amp plays on both channels. The issue is too much hum. I hooked up the reverb earlier today and it works also but as I turn the reverb up the hum increases.

I think the basic design of the amp is ok.

Billy

 

[Tom.G]

As for the HT circuit, the center tap is connected to ground through a fuse. The two other leads are connected to the bridge.

Try disconnecting the CT connection (pull the fuse?) and connect the C38 - C40 junction to chassis.

What is the value of the resistors connected across the DC heater supply, and what does there common point connect to?

 

[Planobilly]

I created a center tap for for the 6.3 AC heater tap at the output tube with two 120 resistors going to ground. So the 6.3 AC is connected at the input of the bridge and also jumps over to the last output tube where the two 120 resistors go to ground. I have all the heaters wired AC at the moment. The way it is drawn in post 19 is the way the 6.3 DC heater circuit has been connected.

Just so I understand, the HT center tap will not be connected and a new connection will be made from C38 junction.

Be back in 3 min

 

[Tom.G]

So the 6.3 AC is connected at the input of the bridge and also jumps over to the last output tube where the two 120 resistors go to ground.

For AC heaters that's fine. For DC heaters you should get better results with the resistors on the DC side of the bridge, at the filter capacitors.

Just so I understand, the HT center tap will not be connected and a new connection will be made from C38 junction.

Correct.

 

[Planobilly]

OK...I understand about the DC resistor placement and will make that change also.

It is 2AM here and I am getting a bit tired. The HT cap in question are a bit of a stacked up mess. I think it better to make these changes tomorrow. I don't want to make some stupid mistake being tired.

I just went back and looked at the schematic in post 6 which I assume you are looking at and looked at the amp. There is a connection to ground now on the negative side of C38. So I assume I just need to pull the fuse? I can take a photo if you like. I can do that now.

 

[Planobilly]

The four turrets in the center are the output of the DC bridge. The green twisted coming out between the caps are the AC wires to the output tube with the two 120 ohm resistors going to ground. There is a black wire to ground from from the cap (C38) The white wire goes to the output transformer

On the back side of the bridge the green and green/yellow wires come from the transformer. The two dark green plastic coated wires go to the pilot light. The cloth covered twisted wires go to the last output tube and the 120 ohm resistors go to ground there.
Perhaps better photos...I hope that helps.

 

[Tom.G]

I expect that fuse from the CT to chassis is blown, and perhaps a diode or two in the B+ rectifier bridge. Just hang your scope on the B+ to verify the ripple is 120Hz. If it's 60Hz then a diode has met its end-of-life.

Yeah, a little past my bedtime too. (11:37pm)

 

[Planobilly]

Yes..fuse is blown. I will check the diodes tomorrow. No 500 volt poking around for this kid tonight!

Thanks you so much for sticking with me. I really needed the help. My internal radar has been beeping about this power supply from day one.

As I have mentioned before, this amp is a bit over my head, a very steep learning curve. But...the only way to learn to fly is by going up in the air...lol

Thanks,

Billy

 

[Planobilly]

Redraw of the possible new HT power supply circuit.

Also the current HT bridge does not have a blown diode as the ripple is 120Hz.

 

[arydberg]

Have you covered the bottom of the chassis with a shielding material like an aluminum sheet? Also is the scope prorerly grounded?

 

[Svein]

So - what is the diode bridge with the zeners doing? One end is connected to the transformer center tap, the other is floating in the air?

 

[rrogers]

Just a couple of general points: make sure the power supply ground is _between_ the filter caps and the rest of the circuit. Otherwise, the bypassed current will be pumping circuit ground.
Always make sure you know where every return path is for every current you use and make sure the forward and return paths/loops have as small an area as possible.
Pretend that the scope ground is 200V@5amps away from the ground of your power supply and hook the scope ground up so that the current will not go through any circuit ground paths. Unless you have a special scope life will be easier.

 

[Planobilly]

Arydberg...Yes and Yes

Svein... More drawing mistakes...sorry...the zener idea is to stabilize the grid supply.

Rrogers...I assume you mean I should remove the first ground I have on the above schematic. I will have to think about your scope comment. I don't have much experience with a scope to begin with...still learning.

Thanks,

Billy