I'm so FU__ING pissed right now! Stupid so called "techs"!

[glpg80]

The wattage you calculate when biasing is nothing more than the amount of power those tubes are wasting as heat.

That would be incorrect.

Guitar amplifiers are rarely single ended or class A by design and always class A/B. If no signal is present on the grids then the current through the plates is the leakage current allowed by the bias voltage on the grid referenced to ground. The total power dissipated of all tubes is the DC summation of all bias currents squared multiplied by the plate resistances of all tubes in parallel. This is still negligible compared to the heat generated from heater supply current for the cathodes to become hot enough for valence electrons to raise states to become more mobile.

If signals are present on the grids then the total power dissipated involves the -3dB efficiency point of the output transformer, reflected impedances, the power factor of the power transformer, and many other variables.

 

[MississippiMetal]

The wattage you calculate when biasing is nothing more than the amount of power those tubes are wasting as heat.

That would be incorrect.

Guitar amplifiers are rarely single ended or class A by design and always class A/B. If no signal is present on the grids then the current through the plates is the leakage current allowed by the bias voltage on the grid referenced to ground. The total power dissipated of all tubes is the DC summation of all bias currents squared multiplied by the plate resistances of all tubes in parallel. This is still negligible compared to the heat generated from heater supply current for the cathodes to become hot enough for valence electrons to raise states to become more mobile.

If signals are present on the grids then the total power dissipated involves the -3dB efficiency point of the output transformer, reflected impedances, the power factor of the power transformer, and many other variables.

My brain hurt reading this. Not because it was difficult to comprehend, but because you went through so much effort to overexplain something I deliberately simplified, and nothing about what you said in any way refuted what I said.

Except for this part

This is still negligible compared to the heat generated from heater supply current for the cathodes to become hot enough for valence electrons to raise states to become more mobile.

If this were true, the valves would not get significantly hotter when the plates are energized, which they do. Sure, the filament wire itself heats up more than the plates, but the plates' larger surface area emits a more detectable heat.

 

[MississippiMetal]

As long as the tubes are decently matched, it's a waste of time checking each one in the 100+ watters. The reason being is in a typical guitar amp with a four-tube power section, it's really two pairs working in push-pull. Electrically each pair behaves as one big valve.

Hence, you only really need to check one tube from each pair, usually one outer tube and one of the inner two.[/quote

Well not really, testing each tube is good because It tells you each socket is working correctly and you haven't blown a screen grid resistor. Also many many times matched sets of tubes are not matched at all. It's good to check. Also you guys really should have a variac also so you can set the wall voltage to 120v or what ever voltage for the country your in. Then bias away. Also don't totally follow the rules listen to the amp at different bias settings. Most 100 or 50 watt el34 or 6l6 amps use a range of 30 ma to 40 ma.. Anywhere in there is fine if you like the sound.

My above statement was assuming the amplifier is healthy and the tubes are matched. If there is an output problem or you aren't confident your tubes are matched then go ahead and check them all.

 

[Racerxrated]

As long as the tubes are decently matched, it's a waste of time checking each one in the 100+ watters. The reason being is in a typical guitar amp with a four-tube power section, it's really two pairs working in push-pull. Electrically each pair behaves as one big valve.

Hence, you only really need to check one tube from each pair, usually one outer tube and one of the inner two.[/quote

Well not really, testing each tube is good because It tells you each socket is working correctly and you haven't blown a screen grid resistor. Also many many times matched sets of tubes are not matched at all. It's good to check. Also you guys really should have a variac also so you can set the wall voltage to 120v or what ever voltage for the country your in. Then bias away. Also don't totally follow the rules listen to the amp at different bias settings. Most 100 or 50 watt el34 or 6l6 amps use a range of 30 ma to 40 ma.. Anywhere in there is fine if you like the sound.

What Dave said. I always go for around 36 for my Marshalls and a little lower with the SLO, around 30. They all sound the best to me in this range. The last Marshall I got was this '87 2205, and it checked in at around 62(!). Dropped it back to 36 and it is one of those rare 'great' Marshalls..they all sound good but this one....

 

[searching]

This work?

https://www.ebay.com/itm/Bias-Tool-probe ... 0392810569

 

[glip22]

This work?

https://www.ebay.com/itm/Bias-Tool-probe ... 0392810569

Get the amp head one. Had two of these and they broke. Amp Head going strong for a long time now.

http://www.amp-head.com/index.php?cPath=21

 

[glip22]

As long as the tubes are decently matched, it's a waste of time checking each one in the 100+ watters. The reason being is in a typical guitar amp with a four-tube power section, it's really two pairs working in push-pull. Electrically each pair behaves as one big valve.

Hence, you only really need to check one tube from each pair, usually one outer tube and one of the inner two.

If the OT sees each pair on the legs, does it pay to match the pairs as close as possible or will they still be imbalance? I test all tubes at the given PV and match the pairs. So , if tube 1-4 tests 36, 32, 33,30, I would pair up the the 36 and 30 for 66MA, and the 33, and 32 for 65ma. As far as the OT is concerned aren't my tubes now matched to within 1MA or will the tubes still be mismatched since the 30 and 36 ma are 5ma out?

No you're doing the right thing. They OT doesn't "see" the tubes so much as they "see" each other and the OT's reflected load. Keep in mind a-a impedance halves when you have four tubes vs two because each pair is in parallel. Because they are in parallel, they operate as one current valve. Between two tubes on one side of the OT primary, the individual values make no more a difference than the individual values of two resistors in parallel. Electrically it's one component. a-a impedance for four valves is only 1/4th of a single tube because each pair is also basically in parallel with the other pair in addition to each tube being in parallel with the valve on its side.

Thanks Bro. Always made sense to me to match them into pairs since the OT has two primarys, not four. Before I understood this I would think if one tube was 6-7 ma out, the set was no good. I have been matching the pairs for a long time now. You are confirming in my above scenario, the OT will see that set of tubes as matched to 1ma as opposed to a 6ma difference from the highest testing for draw to the lowest? Maybe this will help some guys understand why this is the way to go when matching. I know it's not perfect but I take the PV and divide it by the tubes maximum plate dissipation(25 for EL34) Multiply that number by 63 or so percent and bias there. Sometimes I tweak in that range as Dave said to hear the difference. Some amps like it hotter and some not. Accept in a Marshall, or an amp with an EL34 OT primary impedance. That's another discussion as to why a 6550 will not be louder in a Marshall designed for EL34, and biasing it per it's 35 watt MPD is a waste. Treat them as an EL34 since you are limited to the transformer's EL34's impedance which will not allow a 6550 or KT88 to operate efficiently in this circuit. You may not have gone to electronics school for ten years but I know you are a damn good hands on tech.

 

[MississippiMetal]

Most high gain amps I work on, I bias around the 60-65% range of max plate dissipation. Any colder and the tone sounds brittle, nasty and lacking in lowend punch. Crossover distortion is quite audible. If I go much hotter the lowend gets mushy, clarity suffers and tube lifespan becomes a concern. I really only bias upwards of 70% max plate dissipation if I'm working on a low to medium gain amp. Old Marshalls sound great at about 70%, but even then I tend to shoot for around 65.

 

[aw9190]

This work?

https://www.ebay.com/itm/Bias-Tool-probe ... 0392810569

Get the amp head one. Had two of these and they broke. Amp Head going strong for a long time now.

http://www.amp-head.com/index.php?cPath=21

I bought this years ago! great tool to have especially if you have multiple tube amps

 

[searching]

This work?

https://www.ebay.com/itm/Bias-Tool-probe ... 0392810569

Get the amp head one. Had two of these and they broke. Amp Head going strong for a long time now.

http://www.amp-head.com/index.php?cPath=21

Thanks.

 

[LowDesertSludge]

Went to a yard sale that had a few guitars out. Nothing caught my eye, until I noticed a DVD on a table. Tube Guitar Amp Servicing and Overhaul by Gerald Weber. 3.5 hours in and some of this is finally starting to click. This entire thread may as well of been written in Greek before I watched this DVD. Now I understand a little about the components everyone is talking about. Anyone else run through his DVD's or have any of his books? Seems like a great tool.

 

[rusty1070]

Having a bias probe sure beats pulling the chassis out every time you change tubes. good for matching odd tubes too. I use the probe to set the bias then re-check after a couple hours of playing to make sure the bias hasn't drifted into the danger zone.

 

[leib10]

I just got the probe from Eurotubes, and it's served me well for several years now.

 

[mightymike]

Cross over Notch Method :

Put a 1khz Sine wave to amp input ( they have free trial sine wave software online. I just made a wav file of the 1khz sine wave) Some scopes even have a built in funtion generator. I have a function Generator Kit I need to get off my ass and build.
You can get a used dual trace 20mhz scope for $20 on Ebay.

I have a 1/8 mini to 1/4 going from my laptop's headphone jack to the 1/4 input of amp. You could probably use your phone or mp3 player even. Trying to remember how many millivolts I set it to. I hit a solid clean A note on my guitar, measured that, and adjusted my 1khz to that level. Maybe 70 millivolts? I know it wasn't more than 100.

Ok so you have the 1khz Sine going in, amp on, hooked to speakercab or dummy load, then connect scope to speaker jack.

Now you will see the clean sine wave one your scope.

Turn up Amp volume till the peaks start to flatten. This is called clipping or distortion. Go as loud as you can before that flattening starts to happen. Where the peaks are still curved and smooth.

Then you will notice that where the Sine wave crosses the Zero Point, there will be a jagged spike. This is called the crossove notch. When you adjust the bias it will make the lower half and upper half of the sine wave move closer or further apart.

Adjust the bias pot until the notch is gone, or as gone as possible. These are both sides of the class A/B amp like a Marshall or Bassman. The notch just means they aren't aligned. So adust the bias pot till they lined up.

When done, measure bias with a DMM to make sure you are within spec. For example: 6ca7/El34 tube is 25 watt. 70% max dissipation of 25 watt is 17.5 watt. 17.5 ÷ Plate voltage = bias in Milliamp on your bias tool. Or in millivolts accross a 1ohm shunt resister. (Many amps have these so no tool is needed)

99.9% of the time, it is right on the money.. just below 70% max dissipation. I think I was at 68% last time.

You can use the Bias Probe/tool max 70% formula way. And it is just fine. You can put 1 ohm Shunt resisters, and that's fine too, although some swear they affect tone.

But if you are going to use a scope, you need to cross check it after imo.

This is what happened with your tech. I bet he didn't check after doing the scope method, sine wave was too hot, or he just popped them in, and did it by ear, saw they weren't red plating, and just said fuckit.

My friend builds 50 watt class A/B amps with dual bias, external test points, and external access to the bias pots. He can use missmated tubes, and his crossover notch gets eliminated perfectly everytime.

 

[glpg80]

Except for this part

This is still negligible compared to the heat generated from heater supply current for the cathodes to become hot enough for valence electrons to raise states to become more mobile.

If this were true, the valves would not get significantly hotter when the plates are energized, which they do. Sure, the filament wire itself heats up more than the plates, but the plates' larger surface area emits a more detectable heat.

Your oversimplification assumed wasted heat at idle. At idle a class A/B biased amplifier is still wasting energy but only due to heater filaments. At idle the only current in the tube is the sum of the screen grid currents and plate currents at the cathode. Your oversimplification was incorrect and does not take into consideration the true operation of class AB at idle. The reason more heat is produced is because the tube's duty cycle increases when signals are present on the grids. It has nothing to due with surface area of the plate.

 

[MississippiMetal]

Except for this part

This is still negligible compared to the heat generated from heater supply current for the cathodes to become hot enough for valence electrons to raise states to become more mobile.

If this were true, the valves would not get significantly hotter when the plates are energized, which they do. Sure, the filament wire itself heats up more than the plates, but the plates' larger surface area emits a more detectable heat.

Your oversimplification assumed wasted heat at idle. At idle a class A/B biased amplifier is still wasting energy but only due to heater filaments. At idle the only current in the tube is the sum of the screen grid currents and plate currents at the cathode. Your oversimplification was incorrect and does not take into consideration the true operation of class AB at idle. The reason more heat is produced is because the tube's duty cycle increases when signals are present on the grids. It has nothing to due with surface area of the plate.

I never said what the wasted heat was from, and truthfully nobody reading this cares except for you and I. I was saying idle plate dissipation measured is heat, not power dissipated across the load. My point for that was that people look at a datasheet or tube specs on a website and think "Okay it says here that this EL34's max dissipation is 25 watts. This means it's a 25 watt tube" which you and I both know is not the case.