Mesa Boogie Mark 2-B Head, Bias Struggles!

[Platefire]

Got this head in by the same customer that brought me the 74 twin. Only thing he told me on the Boogie is that it has been modded to add a bias pot inside. It had 4 6l6's. Best I can tell from my research is that it is a Mark 3. He just wants it biased. I removed the chassis and found the bias pot. Best I can tell no 1 Ohm resistors has been added.

Next step I wanted to do is fire it up and check the plate voltage and bias voltage. My existing chassis racks don't fit this boogie chassis, so I will have to make a special rack for it before I can do some testing.

He said he wanted it biased slightly hotter than normal. I'm a little fuzzy at this point how to get there without the 1 Ohm resistors in place but I will proceed to build my new rack, check voltages and proceed from there. Any suggestions  would be appreciated. Schematic is attached and seems to match what I got. Platefire

[Willabe]

I don't think that the power amp section in the schemo you posted is correct.   

I think that V6/V7 should be flip/flopped with V8/V9 and the strapped triode should have a resistor between the plate and screen?

And V6/V7 have no grid leak/grid return R's.   
 

[Platefire]

Willabe

Thanks for pointing that out. I looked at the circuitry in the power section a little closer and think it is closer to the Mark 11-B. However there is some small variations but this is close as I can find in Doug's schematics. Check it out and see what you think. Platefire

[HotBluePlates]

He said he wanted it biased slightly hotter than normal. I'm a little fuzzy at this point how to get there without the 1 Ohm resistors in place but I will proceed to build my new rack, check voltages and proceed from there. Any suggestions  would be appreciated.

You can use the transformer shunt method to measure tube current; I don't like the method, but it's handy in this case.

Method:
Turn the amp off. Set your meter to read mA; your meter probably has a ~200mA range, use that. This will likely involved swapping a meter lead into a different jack as well as turning the function switch to a mA setting. Connect the Negative lead to a tube plate; connect the Positive lead to the output transformer center-tap. Clip-on leads with insulated sleeves are greatly preferred here. Now turn the amp on and note the mA reading.

Repeat with the other tubes. I personally turn the amp off or at least on standby when switching tubes. That's because when powered, you will definitely get an arc and an audible pop when you complete the circuit with the meter leads.

You can note whether all tubes seem reasonably matched, and pick the tube you will monitor when adjusting the bias. Connect back to that tube and adjust for desired idle current. Re-check the other tubes to make sure everything is still good.

IMPORTANT: When done biasing, stop and move the red meter lead back to the volts jack. Switch the meter function switch back to volts, and you're ready to use your meter for other tasks. If you're not deliberate in doing this, you'll forget and go to measure a voltage with you meter leads connected for current measurements. That makes your meter act like a short circuit, could damage an amp component, and if you measure B+ voltage somewhere will almost certainly pop an internal fuse in the meter.

I didn't note the "Important" step because I've popped 4-5 meter fuses being dumb and not setting the meter back up for voltage measurements. Nope, I never made that mistake at all. Never had to go find obscure fuses to get my meter working again...

 

[Willabe]

Check it out and see what you think.

Sorry Plate, I don't know, you have the amp.   

That schemo's power amp section looks like it's right? 

Check to see how many wires come off the OT's primary. For a normal PP amp should be only 3 wires, CT and PP ends. For an OT that's UL, that Mesa use's for their Simal Class amps, should be 5 primary wires.   

[Platefire]

Thanks HBP

I have done that shunt method but it has been years. Thanks for reminding me. As I recall you have to be super careful with this method or you can do major damage. I'll have to do a personal seminar refresher course to get back up to speed.

Regarding the Mark 2-B schematic the difference from the schematic circuit and actual circuit is:

a-On "B" power node that feeds power tube screens, it has an extra 10K/2 watt resistor prior to all the 470 Ohm screen resistors.

b-On the two outside 6L6 sockets, there is what looks like a diode going from the plate pin to the cathode pin. I think it is a diode because it has a single ring around it on one end.  The ring is on the plate pin side.

c-The two 6L6 sockets on the right side/or top side of the schematic had a 50 Ohm resistor to ground on pin #2. The other two 6L6 sockets don't have that.

These are not on the schematic but on this amp. Other than that, everything immediately around the power tubes is the same. Platefire

[Platefire]

Ok, got my chassis rack build for the boogie. Looks like the voltage on the Power amp section is running about 50V above the schematic. The schematic don't show plate/screen voltages on the power tubes but the power node to PT is 445V. Bias voltage on sch is -51. So here are the voltage readings:

Plate/Screen
 497/498
 500/500
 501/499
 501/497

Bias Voltage=-43

Again most boogie's the bias is permanently set but on this Mark 2-B someone modded it to add a bias pot. Here is a pic of the power tube arrangement and bias pot. Guess I'll get up to speed on the shunt bias method. Platefire 

BTW-Fired up the amp on rack while ago to get voltage readings and to get prepped for biasing. The tubes were not red plating and temperature on glass felt about right--hot but not too hot.

[Platefire]

OK, using the shunt method, my Fluke 115 says between 4 new looking Ruby 6L6GC's, one side is 66 mA(33 mA per tube) and the other side 72 mA (36 mA per tube).

As usual the tubes are not too close as far as being balanced.

He said he wanted it biased on the hot side of normal or slightly hot. If 35mA is considered normal--With the existing unbalanced situation my idea would be to move the hotter side up to about 39 mA and that would move the colder side hopefully to the normal range. Any suggestions?? Platefire

BTW-Webers Bias calculator says A/B amp with 6L6GC's operating at 500V plate voltage should be biased at 42 mA. Would I be safe to go that high adjusting to the hot side???? 

[Willabe]

3mA difference from push side to pull side is plenty close.

But you have 4 tubes and your assuming both tubes per side are matched too. Maybe yes, maybe no.

I'd check them separately from each other just in case. Put them in a 2x6L6 amp and take a reading for each tube.

Then you'll know for sure that 1 tube is not pulling more current then the others but you don't see it because the tube it's paired with is weaker and pulling less current.   

[2deaf]

This amp has the two-tube or four-tube output tube option.  Put it on half power to test single tubes instead of pairs.

[Willabe]

This amp has the two-tube or four-tube output tube option.  Put it on half power to test single tubes instead of pairs.

Good catch! 

[Platefire]

Well I'm officially cornfused now!

My previous test was with the 1/2 power switch on. After the previous post by 2deaf I went and checked it and sure enough, it was set to 1/2 power. So I did another test full power. The hotter side read 121 mA and the cooler side read 115 mA. Both of those divided by two 60.5/57.5. I don't get it unless that needs to be divided by 4???

What am I reading here? Platefire

BTW-This is a customer's amp, same one who brought the 74 twin(another thread)

[Platefire]

Bump! Can somebody please explain my last ma readings! I can't proceed until II determine what last readings mean?

[Willabe]

My previous test was with the 1/2 power switch on. After the previous post by 2deaf I went and checked it and sure enough, it was set to 1/2 power. So I did another test full power. The hotter side read 121 mA and the cooler side read 115 mA. Both of those divided by two 60.5/57.5. I don't get it unless that needs to be divided by 4???

Well if the 1/2 power switch lifts the K ground connection on 1 pair of power tubes then those 2 tubes are off.

So you measured with 2 power tubes off;  "one side is 66 mA(33 mA per tube) and the other side 72 mA (36 mA per tube)".

Then; "So I did another test full power. The hotter side read 121 mA and the cooler side read 115 mA."

That should be all 4 tubes. And it looks right because 4 tubes mA's = about 2x of the 2 tubes mA's.

Looks like in full power mode the 4 tubes run a little cooler than in 1/2 power mode. Probably sightly different dc plate voltage in 1/2 power mode, a little higher? And in 1/2 power mode you eliminate the UL taps, so OT primary see's slightly different reflected impedance?

[Platefire]

Thanks Willabe. You'll have to forgive my ignorance on this. Haven't quiet dealt with it this way before

Yes I checked with my MM the continuity on the cathode to ground and it is broken on half power on the two inside tubes.

Also checked the plate voltage in both modes and full is about 488 and half is 501.

It seems to me with the voltage drop of full power, getting the bias right for both modes is somewhat of a balancing act. So tell me this, is my thinking right or not----if the existing average Highest mA is 36mA in half power mode and highest average mA in full power mode 30.25mA it seems if you adjusted the bias in half power mode where the 36mA average is increased to 40mA--being the peak setting and that would bring up those other lower average mA up also approximately 4mA higher???? Whatayouthink? Platefire

[2deaf]

It sounds to me as though each tube is idling around 60ma, not 30ma.  What happens if you cool them down to 40ma or so?

These are not PCB octal sockets, so you can easily raise one cathode pin and put a 1 ohm resistor in. 

[Platefire]

Well I fired it up and played my guitar over it a while just wanting to see how hot the tubes were getting. Ididn't detect any red plating. I can tell you they are getting very hot to the touch. It don't take long to look at a hot potato I decided to get a temperature reading with my Infra Red thermometer. The hottest spot on each tube seems to be right at the top edge of base where I'm getting 300F+ readings. This seems way to hot compared to previous readings I've done on other amps. Platefire

[2deaf]

-43V for bias with around 500V on the plates is way too hot for any 6L6 with any life in it.  The -51V that Mesa allegedly notes is a much more typical value for these conditions.  Give it a try and see what the idle current is.

[2deaf]

What makes the customer think that the tubes need to be biased hotter?  Maybe he is just unhappy with the way that it is biased and really what he wants is to have the tubes running cooler.

Those IR thermometers are always fun to point at a power tube, but I don't think it is a valid test for much of anything.

[Platefire]

I didn't know the status of the amp when he brought it in. He didn't indicate anything was wrong with the amp. He just said he wanted it biased. My interpretation was that he wanted me to check it to see if it was biased correctly and adjust it accordingly. So I ask how he wanted it biased, normal(not cold or hot) or on the hot side. He said the hot side--which I interpreted is the hot side of normal. So if for a 6L6GC normal is considered 35mA the hot side of normal would safely be in the range of 36 to 40.

I will try an adjustment tomorrow to bring the high side(72mA in 1/2 power mode) to around 40mA and see where the bias negative voltage is reading at that point. Thanks, Platefire

[Willabe]

You'll have to forgive my ignorance on this. 

Ha, are you kiddin me? I'm just guessing. 

Yes I checked with my MM the continuity on the cathode to ground and it is broken on half power on the two inside tubes.

Smart! 

[sluckey]

You need to consider more than just the current through the tube. You also need to know the plate voltage. Now multiply the plate voltage times the current to determine the static power dissipation. For the tube that has 72mA and 500v, the power would be 500 x .072 = 36Watts! That's way too hot for a 6L6.

Max plate dissipation for a 6L6 is 30 watts. Most people like to set the bias for 70%. For a 6L6 that would be .7 x 30 = 21W. I would call that the norm. 21W/500V = 42mA. So, adjust the bias so that you have 42mA flowing through the tube that is presently showing 72mA. then recheck the plate voltage and do the math. You may have to juggle the bias setting to come up with exactly 21W. See how the amp sounds at this setting. Adjust hotter if desired, BUT DO NOT EXCEED 30W! (60mA X 500V). Everytime you adjust the bias pot, recheck the current and plate voltage for all four tubes.

I urge you to do as 2deaf recommended, ie, lift pin 8 from the board ground and connect a 1Ω resistor between pin 8 and ground. Do this for all four tubes. Now you can safely and quickly/easily measure the current thru every tube without having to shut down to move probes. (If I had customers I'd also keep a bag full of 1Ω 1W 1% resistors on hand. I'd also build four of Doug's bias checker adapters and install 1Ω resistors on each. Using these modified adapters allows you to easily check bias without modifying the customer's amp)

There's a way to determine the exact current flowing thru each tube without having to divide by two which ASSUMES that each tube in a side is equal (rarely true). The half power switch makes this possible. You simply measure current at full power then measure current at half power. Subtract the half power numbers from the full power numbers to get the numbers for the other two tubes. For example, these are your numbers for half power and full power...

you measured with 2 power tubes off;  "one side is 66 mA(33 mA per tube) and the other side 72 mA (36 mA per tube)". This is for two tubes

Then; "So I did another test full power. The hotter side read 121 mA and the cooler side read 115 mA." This is for all four tubes

So, 121 - 72 = 49mA and 115 - 66 = 49mA. These are the numbers for the other two tubes. 66, 72, 49, and 49 are the numbers for all four tubes.

[Platefire]

Thanks very much Sluckey for the help!!!

I re-set the bias using the highest previous 72mA half power side and adjusted it to 42mA. The new bias voltage at this setting is -53.48(sch says -51). This changed the plate voltage in half power mode to 511.5VDC plate voltage.

Measuring the other p/p side in 1/2 power mode with this bias setting, now 38mA and 509.6 plate voltage.

In full power mode the high side of the p/p now reads 73mA & 504V plate voltage and the other side 67mA and 505.3V plate voltage.

Doing the subtraction you referenced in previous post, full power reading on hot side 73mA-42mA(1/2 power reading)=31mA for other tube. On cooler side 67mA-38mA=29mA for other tube. Platefire

BTW-as a second thought on the above reading, guess it wouldn't hurt to move the high side up to 50mA and then re-measure everything. That should bring the low one at 29mA up a bit hopefully around 35mA.

[2deaf]

Now try moving the two inside tubes to the outside and the two outside tubes to the inside and try it again. 

Sorry I misunderstood about the customer wanting a hotter bias.

[Willabe]

Now try moving the two inside tubes to the outside and the two outside tubes to the inside and try it again. 

+1

[Platefire]

Did Tube swap as suggested. New reading are as follows:

Left p/p side=Half power=42mA/505 plate voltage/-50 bias voltage
                    Full power=81mA/494 plate voltage/-50 bias voltage


Right p/p side-half power=42mA/505 plate voltage/-50 bias voltage
                     Full Power=84mA/494 plate voltage/-50 bias voltage

So the p/p sides are averaged out: left side 81ma-42mA=39mA. So we have a 42/39 mA combination
                                                   Right side 84mA-42mA=42mA. So we have a 42/42 combination

Fellows, I don't know about you but I think we have a correctly biased amp. Guess it could go a little hotter but I think this is fairly safe for the tubes. WhataUthink? Platefire

[DummyLoad]

Did Tube swap as suggested. New reading are as follows:

Left p/p side=Half power=42mA/505 plate voltage/-50 bias voltage
                    Full power=81mA/494 plate voltage/-50 bias voltage


Right p/p side-half power=42mA/505 plate voltage/-50 bias voltage
                     Full Power=84mA/494 plate voltage/-50 bias voltage

So the p/p sides are averaged out: left side 81ma-42mA=39mA. So we have a 42/39 mA combination
                                                   Right side 84mA-42mA=42mA. So we have a 42/42 combination

Fellows, I don't know about you but I think we have a correctly biased amp. Guess it could go a little hotter but I think this is fairly safe for the tubes. WhataUthink? Platefire

ship it!   


--pete

[Platefire]

Yeah, that's what I say, ship it! I am letting it run about 30 minuets in full power mode and another 30 in half power and watch the plates but if they weren't red plating before, don't think they will now but just giving them a run through to make sure.

HotBluesPlates, Willabe, 2deaf, sluckey----thank you very much for helping me with this!!! I will dance at your next wedding! Platefire

Oh yeah thanks DummyLoad for putting the seal on it