trinityamps.com

Tube Amp Troubleshooting

Take voltage readings of your tubes in case someone on the forum needs to see those readings.
Voltage readings are like heart beat and blood pressure readings that your doctor uses to diagnose your health issues.
The tubes need to be in the amp, the amp needs to be on and the power tubes biased properly to get any sort of meaningful voltage readings


For the most part, an amp is basic triode gain stages put together in different sequences.

The checklist below doesn't cover everything but it is something to always bear in mind, whether you're fixing an existing amp or troubleshooting your own build.

With tubes installed, measure and record all the voltages and compare them to whatever documentation you have. High plate voltages or low cathode voltages indicate the tube is not conducting properly. Could be a bad plate or cathode resistor, poor ground, bad bypass capacitor or vacuum tube.

Use FX loops if present to isolate a problem to preamp or PI/output. Then is the noise affected by a MV? Tone stack? Channel volumes?

Pull preamp tube 1, turn on amp and see if problem has gone. If so, focus in this area. If not resolved, pull tube 2 and repeat. This will help isolate the location of preamp issues. If the problem goes away, try replacing that tube.

Once you narrow it down to a single triode or maybe two possible triodes, here's a checklist of things a triode needs to function, aside from making sure it's not just a bad tube:

1) Plate resistor.

2) Cathode resistor, bypassed or not. The bypass cap is always suspect.

3) Grid leak resistance. This may be a volume pot, a fixed resistor, or a tone stack. No matter how it's accomplished, the grid must have a connection to ground for current to flow. And ensure the grid leak is always there, even if a switch connected to the grid is engaged ("It pops when I change channels")

4) In AC coupled triode stages (the vast majority) make sure no DC is leaking from the previous stage to the grid.

Confirm that 1-3 are connected and measure ok with an ohmmeter.

Measure 4 with a voltmeter.

Measure voltages at plates and cathodes. If there isn't a big voltage drop across the plate resistor, the tube isn't conducting. One of the four issues above is the cause. Even if they all check ok as far as measured resistance goes when the amp is off, something isn't passing current. For example, it could be a bad ground connection not allowing current to flow so voltages are way off, such as very high plate voltages.

With no HT voltage (completely discharged safely) , all anode resistors need checking and cathode capacitors (25uF 50V etc) might require replacing.


Other Tests
Check all ground connections to ensure they are solid, clean and conductive. Measure resistance <1 ohm.

Noise or cracking.

Crackle could be caused by any resistor. Grounding grids or pulling tubes lets you find the stage where the crackle is.

First track down where the crackle is coming from. Either by grounding grids or pulling preamp tubes, working from input to PI. Pull tubes starting with the first preamp tube until the noise goes away. That narrows down the problematic area. Try replacing the tube first before diving in.
Check value/integrity of plate resistor.

Then it's a combination of the above list, the way the amp was constructed, and materials used. In eyelet/turret/tag/terminal builds, solder joints are pretty easy to inspect. Touch up suspect solder joints by reheating and adding a little solder.
Look for discolored components. Smell things.
Measure any voltage from the board material itself to ground!

LEAD DRESS

Try to maintain a linear path for audio. Think of the signal levels in each wire as you route them. Keep inputs or low level signals away from high level wires. When in doubt, use shielded wire.

Twist heater wires. What is really important is the distance of the heaters to the input grid wires.
You can fly the heater wires in the air & away from the chassis and place the the grid wires against the chassis. Or if you wish to go the other way fly the grid wires in the air & away from the chassis and place the the heater wires against the chassis in the corner.

Don't have your OT's input and output wires run parallel to each other in close proximity for anything more than is unavoidable. And the reason is RF interference via induction. Twist the input and output side for common mode noise rejection. Even a few turns will help quite a bit.
The closer together they are, and the longer the run that they are together, especially with no twist, will lead to noise.

If you must cross a signal wire with another wire try and cross it at 90-degrees to minimize inductive interference.

Keep high signal lines away from low. If you must cross them, do it at 90 deg. Every lead has a magnetic field. If you have an alternating current in one it can induce a current in an adjacent wire. Parallel lines make this worse, 90 deg should minimize it. Same as the transformers; they’re nearly always 90 degrees out (on their cores).

Identify which leads carry AC and which DC and keep AC runs short and let DC runs be longer. Audio signal is AC.
Twist all AC wires if possible for as much of the length they run as possible; even a few turns cuts down RF noise (common-mode-suppression) Power leads, rectifier and heater leads and remember: your OT wiring is AC as is any wiring running to, or from, a transformer, and that includes the choke.

Typical amp has short DC path from tube to tonestack components then three long AC paths to the pots. If you have the components right by the pots then you have a single longer DC path from the tube and short AC paths.

NOTE - Measuring "continuity" is NOT a guarantee of determining a short. The "continuity" setting on a multimeter typically generates a tone when the measured resistance is 199 ohms and less. That last statement is very, very important to understand. As an example, if you are measuring the continuity of a wire or switch contact or ground connection and a tone is created, this indicates that the resistance between the leads is somewhere between 0 ohms and 199 ohms. That is potentially NOT a short. For an accurate determination of a short, you'll have to switch to a resistance setting and measure the actual resistance.

TIP: Sometimes a larger diameter solder is needed (especially for ground connections). Instead of buying an entire roll, a quick solution is to just twist together two strands of the smaller stuff. A simple rolling between the fingers is usually sufficient to get a tight twist.

Note: If you have a suggestion, let me know and it can get added to this .

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps