How do I modify this bias circuit Mesa Nomad 45 tube amp

[Planobilly]

Hi guys,
I was given this Mesa Boogie Nomad 45 chassis. I repaired it yesterday and have it working more or less well.
Mesa does not install adjustable bias in any of their amps. I want to install a trimmer pot to control the bias.
Where in the circuit should I place the trimmer resistor and what do you think the value of the trimmer resistor should be?
http://schematicheaven.net/boogieamps/boogie_nomad45.pdfCheers,

Billy

 

[Svein]

I want to install a trimmer pot to control the bias.

Why? Valves do not require any bias (with the proper cathode resistor they are self-biasing).

 

[Planobilly]

Why? Valves do not require any bias (with the proper cathode resistor they are self-biasing).

Hi Sevin,
The simple answer to your question "Why" is I want control over the plate current. The actual answer is much more complex.

All valves require bias and there are three basic ways of doing that.

• Non-Adjustable Fixed Bias amplifiers (no bias adjustment potentiometer)

• Adjustable Fixed Bias amplifiers (adjustment potentiometer to set the "fixed" bias point)

• Cathode Bias amplifiers (also commonly referrred to as self-biasing

Cheers,

Billy

 

[Svein]

The reason self-biasing is used with valves is that the anode current is dependent on several factors, such as the age of the tube, the filament current, the voltage at the screen grid etc. Anyhow, the schematic you linked to is just a block schematic, it does not show the actual circuit.

If you really want to control the anode current, I would suggest varying the screen grid voltage. Almost no current enters or leaves the screen grid, so you can use a high-resistance potentiometer. Due to the voltages involved, you cannot use a standard trimmer, however.

The obvious place to insert a trimmer would be at the bottom of the grid resistor, but I do not think that you have access to a negative voltage. There are other ways, but most of them involve a thorough understanding of valve amplifiers - and that understanding is scarce nowadays.

 

[Planobilly]

Hi Svein,

The schematic is kinda of a mess but if you page down to page 7 you will see the power amp section and the 4 El84 power tubes. All the way down at the last page called page 2 you will find the power supply schematic which contains a -15v dc supply.

The grids are currently bias per the schematic at -12vdc and actually measuring -11.5Vdc. Very small changes in the negative voltage results in large changes in plate current. Point being I want to be able to change that negative voltage up or down as needed within the limits of the negative supply voltage that exist.

To make the issues even more complex EL84 tubes/valves have per data sheet a max plate voltage of 300V with a plate dissipation of 12 watts. It is very common for designers to run the plate voltage up to 400 plus volts as is the case in this amp. This makes the tubes run hot as hell but they sound good. This is OK if one is looking for a really loud amp and you are willing to put up with reduced tube life. I am not wanting a loud amp.

JJ, a current MFG of tubes has designed a new EL84 style tube called a EL844 which dissipates 9 watts instead if the normal 12 watts. Installed in this amp it will reduce the volume a good bit but will break up much faster or stated another way go into distortion sooner. I have these tubes on order and they will arrive in a day or two.

None of this bias business means anything unless the tubes are well matched, meaning the trans-conductance values are close for each tube.

I am sure all this vacuum tube stuff makes modern day electronic engineers a bit crazy but vacuum tubes are not going away anytime soon and not just as a result of a bunch of guitar players. There are things that tubes do that solid state devices just can not be made to do a this point...perhaps in the future...who knows.

Cheers,

Billy

 

[jim hardy]

I think i'd add a few hundred ohms of cathode resistor to each tube, bypassed with a big electrolytic capacitor of course
and make a master bias control so

that'd give you 5volt adjustable bias, only adding a milliamp to power supply load
rheostats in the cathodes would make the tubes individually adjustable
Size for maybe 5 volts at nominal cathode current ?
maybe a half the R adjustable and half fixed ?
Just bypass them so as to keep gain same.
A voltmeter mounted on panel with a switch to read across each tube's fixed part of R_cathode would let ypu balance them.

have fun !

old jim

 

[Svein]

The schematic is kinda of a mess but if you page down to page 7 you will see the power amp section and the 4 El84 power tubes. All the way down at the last page called page 2 you will find the power supply schematic which contains a -15v dc supply.

Sorry, I had not seen that part - and I wish I hadn't. I have 15 years of experience with valve amplifiers and this looks absolutely crazy. The EL84 was a common output valve for a 3-5W amplifier, maybe 10W in a push-pull setup. It was never the valve you would use for a high power amplifier (that would be EL34 or 6L6). And running way beyond spec...

 

[jim hardy]

EL84/6BQ5 was common in "good" 1960's hifi gear
in high school i built a stereo amplifier , used a pair for center channel subwoofer .
They'll make 15 watts push-pull if you don't mind some distortion.

 

[Svein]

They'll make 15 watts push-pull if you don't mind some distortion.

Usually, we do mind distortion, at least in the level your figure shows. 3-4% distortion - even in the 60's that was a lot. The "golden standard" was the Williamson amplifier (https://en.wikipedia.org/wiki/Williamson_amplifier) which had less than 0.1% distortion - in 1947!

 

[Planobilly]

Hi guys,

Sorry, I had not seen that part - and I wish I hadn't. I have 15 years of experience with valve amplifiers and this looks absolutely crazy. The EL84 was a common output valve for a 3-5W amplifier, maybe 10W in a push-pull setup. It was never the valve you would use for a high power amplifier (that would be EL34 or 6L6). And running way beyond spec...

I had to laugh a bit when I saw your post...lol This is a crazy amp! It has 23 potentiometers to control the sound all of which were just hanging off the PCB when I got the amp...lol It was a bit of a job just to find where they all went! I think one would need a degree in physics from MIT just to turn the bloody thing on..lol

The distortion issue for guitar amps in general will make little sense to a electronic engineer unless he is also a guitar player/musician. Amps for sound production are one thing and amps for sound reproduction are another.
The design of this amp makes little sense from a engineering perspective. The reason for the use of the El84 is purely about "tone" and also how the guitar responds to the amp. This is a "system" defined by the guitar player/musician/ the guitar, the amp, and the speakers. It is hard to discribe to someone who does not play guitar but in general one does not play the guitar in the same manner with different styles of amps. The "tone' of the amp is in large measure in your fingers and how you touch the guitar strings combined with the amp design and how the controls are set. Many guitar players are on a truly "insane" quest for tone..lol

El84 amps are very common and well liked by many guitar players, perhaps less so in push pull forms than in class A. There is just nothing that sounds like a well designed class A El84 amp. Bruce Zinky a designer who worked for Fender for many years and now has his own company is a classic example of a great EL84 class A designer. He is not a household name for sure but well known by all of us into amps.

When I got the amp I only had four EL84 tubes in stock, one matched set of two and two others. The plate voltage was 400V on all the plates. The grids were at -11.5v and shown at -12v on the schematic. The plate current was running at 35.2, 35.4 33.8, and 38.0 ma. There may be other things to account for the high plate current that I have not discovered yet as I have not measured every value of resistors and caps. I would assume something around 25ma would be more normal but I don't really understand the effect of running the tubes so far out of spec. The tubes are running at 265F which is hot but as there is no max temp spec I can find on a EL84 data sheet I am not sure just how hot they can run. I know the ones in my Zinky 30 watt amp run hot as hell. A 6l6 for example can run at a max of 450F per the data sheet. I would never run one that hot but I guess one could.

All this kicked off the question of changing to Adjustable Fixed Bias. The new quad matched set of JJ EL844 will be here Monday so I will see how they run, and make a decision at that time of what to do about the bias change mod.

Again guys, this is a very new adventure into the world of electronics for me so there is a ton of stuff I do not understand yet. The learning curve looks like a pretty tall mountain from where I am standing at the moment.Cheers,

Billy

 

[Svein]

All this kicked off the question of changing to Adjustable Fixed Bias. The new quad matched set of JJ EL844 will be here Monday so I will see how they run, and make a decision at that time of what to do about the bias change mod.

OK. Look at the schematic I copied into post #9. There you see a couple of potentiometers to trim the anode current and the balance between the two output valves. Also observe that they are specified as 100Ω, 2W wirewound. That is not a small trimmer, that is a large ceramic-based thing.

Also, 35mA no-signal current is not too bad - for a class A or AB amplifier. The minute you start to think about class B, the distortion increases significantly.

The tubes are running at 265F which is hot but as there is no max temp spec I can find on a EL84 data sheet I am not sure just how hot they can run. I know the ones in my Zinky 30 watt amp run hot as hell. A 6l6 for example can run at a max of 450F per the data sheet.

Yes. You are using electronic valves (or "tubes" as you call them). That is basically an incandescent lamp with some metal inside. The heater draws 760mA at 6.3V, so you have 4.8W dissipation just there. Add the 400V/35mA standby anode current and you have another 14W dissipation. So, before you introduce any signal, you are burning almost 20W - and a 20W incandescent bulb gets hot!

 

[Planobilly]

This is a general chart that is widely used and 70% of the total value being a safe and sort of normal value. These values shown on the chart agree pretty close to the MFG data sheets I have seen. There are current MFG's that are producing 6l6's that indicate 30W dissipation for example but all the MFG data sheets for EL84's indicate 12W

So 12W/400V X .7= 21 ma If the valves in this amp were running at max data sheet stated voltage of 300V then the 70% value would be about 28 ma.

I have always assumed that the wattage given in the chart indicated max signal power output. So perhaps I really don't understand what max signal power means. The 4.8W dissipation from the heater just escapes as heat correct?

Here is the JJ EL84 data sheet I am basing my info off...sorry I don't know how to cut and paste a PDF. http://www.drtube.com/datasheets/el84-jj2003.pdf

So perhaps you could explain why you consider 35ma to be "not to bad" in the context of the EL84 running at 400V.

I am a bit confused...as usual..lol

Cheers'

Billy

 

[Svein]

So perhaps you could explain why you consider 35ma to be "not to bad" in the context of the EL84 running at 400V.

Plain and simple: 15 years of working with EL84 and similar valves. And - as I said - you are essentially running a class A (or possibly AB) amplifier (http://sound.westhost.com/class-a.htm), which means that you will have a relatively large no-signal current.

 

[jim hardy]

That datasheet snip in post 8 is from the RCA tube manual at the link you posted. I copied the pdf into my 'electronics' folder, all700 pages...

Observe it says~60 to~70 ma for idle current. (did i snip right page?)

sorry I don't know how to cut and paste a PDF.

Windows has a handy "snip" tool , try it out... snip, click edit, click copy, open Paint and paste, add your notes and save ( i use jpg).
A PF mentor showed it to me.

 

[Planobilly]

Thanks Svein,

I read the link you provided. Need to go back and study it a bit.

Cheers,

Billy

 

[Planobilly]

Hi Jim,
Yes, the one you posted matches my RCA book. If I understand what I am reading Maximum Signal Power goes up as plate voltage increases per the RCA book...11W at 250V and 17W at 300V so 400V...?.This is the JJ data sheet for the tubes I have in the amp at the moment..

This one below is from a Russian Data Sheet...the first one I have seen that gave a max value for max temp. 573K which =571.73F...WOW! that's HOT..but Borosilicate glass has a melting point around 820oC (1,508oF) so I guess that could be correct.

If you have the time take a look at this from Aiken Amplification and see if it all seem correct to you. http://www.aikenamps.com/index.php/the-last-word-on-biasing

Thanks,

Billy
EDIT; Another Aiken link with graphs http://www.aikenamps.com/index.php/idle-current-biasing-why-70-percent

 

[jim hardy]

If you have the time take a look at this from Aiken Amplification and see if it all seem correct to you.

a quick read looks very logical and common sense.

I like cathode bas because it's self restoring, ie negative feedback. "Always encourage Mother Nature help your design work..." a wise old engineer once told me.

You have the opportunity to blend a little bit of cathode bias in with your grid bias. Individually adjustable cathode resistors would let you match individual tube plate currents, as well as giving a handy spot to measure cathode currents. A galvanometer between cathode resistors would null when they're matched. If this is a 'showoff' amp, imagine your friends' faces when they see you can match idle current at the start of a session.

Have fun !

 

[Planobilly]

Hi Jim,

Thanks for the feedback.I found a bunch of useful info on the Aiken site. There was a part on safety and this was one tip I had not considered...lol

• Never work on amps around small children - you will be embarrassed when they pick up the language you use when you touch live wires.

Have a good Sunday

Billy

 

[f95toli]

Sorry, I had not seen that part - and I wish I hadn't. I have 15 years of experience with valve amplifiers and this looks absolutely crazy. The EL84 was a common output valve for a 3-5W amplifier, maybe 10W in a push-pull setup. It was never the valve you would use for a high power amplifier (that would be EL34 or 6L6). And running way beyond spec...

It is a Mesa Boogie guitar amp, it is supposed to have lots of distortion...
Mesa Boogie amps are used by the likes of Carlos Santana and Dave Grohl, i.e. they are mainly used for a "heavier" sound.

 

[Svein]

It is a Mesa Boogie guitar amp, it is supposed to have lots of distortion...
Mesa Boogie amps are used by the likes of Carlos Santana and Dave Grohl, i.e. they are mainly used for a "heavier" sound.

Then why go to all the trouble with a tube amplifier? Introduce most of the distortion in a FET pre-amp and use a class D end stage...

 

[Planobilly]

Update

The basic amp is running. There are several things that are not working.
No foot switch or DIN cable. Ordered and in transit...used foot switch from the internet at $65.00...new 5 pin DIN cable from Mesa $27.00
No reverb tank and had to order from Mesa at a high price of $50.00 (I do not know if the reverb circuit is working...will check when the tank gets here)
There is no output to the head phone jack. There is a NE5532AP opamp which may be the issue. Data Sheet http://www.ti.com/lit/ds/symlink/ne5532.pdf
There seems to be some switching issue but when I get the foot switch I will have a better idea of what is going on
The schematic I have is not the best but Mesa is emailing me what they have...perhaps better perhaps not.

I have done some experiments using EL844 tubes which so far I don't like the sound of. The idea was to use a lower out put tube which would reduce the overall volume and cause the tube to break up quicker. This works but I don't like the tone. This tone issue may be a function of bias issues with these lower out put tubes.

At this point I am trying to get the amp back to factory specs and after I will start on the process of making mods to the amp.

Because I got the chassis for free the cost are not out of control...well not yet at least..!..lol Looks like around $300.00 by the time I build a case.

Cheers,

Billy

 

[Planobilly]

Hi Guys,

Jim an I have done a bit of troubleshooting on this amp. So perhaps I should bring everyone up to date.

Disregard the original questions about bias and bias issues. All the tubes are running at normal voltages and all three channels are working more or less. I now have a foot switch and it is working normal.

This is what is not working.

The reverb unit is not working. There is no negative voltage at the op-amp which is a NE5532 and no negative voltage at the two Zener diodes for the low voltage supply.
The record out/headphone circuit is not working.
Channel two master volume control is not working and when I switch to channel two the channel three master volume is controlling channel two. My guess is that is a relay issue. Also the channel three LED is always on which I assume is not normal.

I have ordered several parts from Digi-Key which will be here on Wednesday to include new relays so some things can not be done until then.

My question for today is can I disconnect the record out circuit all together for the moment and can I disconnect the reverb circuit. I need to be sure that whatever is causing the negative supply to fail can be isolated .

I ASSUME that removing the .047 cap in the record circuit would disconnect the circuit without issue. YES...NO...Better idea?

As for disconnecting the reverb circuit I am not sure?

As you can see there is a ton of stuff crammed into this amp. Also there are traces on both sides of the PCB and very little information on the PCB that relates to the schematic. This is one of the most complex guitar amps I have tried to work on.

Cheers,

Billy

 

[jim hardy]

My question for today is can I disconnect the record out circuit all together for the moment and can I disconnect the reverb circuit. I need to be sure that whatever is causing the negative supply to fail can be isolated .

Record and Reverb opamps are capacitor coupled, so disconnecting them won't hurt anything.

I ASSUME that removing the .047 cap in the record circuit would disconnect the circuit without issue. YES...NO...Better idea?

That'll disconnect signal
but not power.
If something is overloading the -15 volt supply , what clue might give it away ?
It might be warm to the touch... the pcb trace bringing it -15 might drop more microvolts per inch...

 

[Planobilly]

Thanks Jim,

Yes, I guess disconnecting the signal does not do us any good.

Well, as much as I am not looking forward to the job I guess I have to bite the bullet and remove the PCB and check everything very close. There are several places where parts are so close together there is less than the width of a business card between the solder rings. As there are traces on both sides of the PCB one can not see what goes to what with the PCB in place.

I have to do that in any event to change relays.

Cheers,

Billy

 

[Planobilly]

Update

I have the low voltage supply working now.
I fixed the record out circuit by replacing the op-amp with a new original TI NE5532
-15V and +15V to the both op-amps now.

Reverb still not working. I need to replace the general purpose NE 5532 op-amp I put in with the new Texas Inst. original I now have. I am not sure if that will help but at least I have the correct voltages going to the op-amp now. I may do some signal tracing before I change it.

Channel 1 / Channel 2 switching is not working correctly.
The NEC relays I ordered from Digi-Key are not the proper relays. The original ones are marked NEC MR62-6USR 0.6A 110VDC...2A 30V DC and 1A 125VAC. Digi-Key could not find those numbers and sent general purpose that they thought would work. I can not find a data sheet under those original part numbers. I may be able to get the originals from Mesa. In fact , there may be nothing wrong with the relays as it relates to channel 2/channel 3 switching issue as it could be a fault in the LM3914N or somewhere else in the switching matrix. Relays in guitar amps in general are always going bad so I always replace them while I have the boards out. Cheep insurance!

Well...midnight thirty here...tomorrow is another day!

Cheers,

Billy

 

[jim hardy]

http://www.mouser.com/ds/2/283/mr62-303636.pdf

this looks like equivalent
https://www.omron.com/ecb/products/pdf/en-g5v_2.pdf

http://www.digikey.com/product-detail/en/omron-electronics-inc-emc-div/G5V-2-DC6/Z3280-ND/369044

 

[Planobilly]

Thanks Jim,

I been in Miami all day and just got home. I am going to do some testing now to see if I can find the channel switching fault. Then I want to change the other op-amp and install a socket while I am at it.

I am down to only two issues left. The reverb and the channel switching issue.

Cheers,

Billy

 

[Planobilly]

Udate
I changed all the relays except one which I forgot to order the correct amount. I need one more which is coming this week. Having said that, I bench tested the relays that came out of the amp and they all working. They may or may not be switching correctly. My hat is off to anyone who can understand the switching matrix in the schematic.

The red LED for channel 3 is always on. I traced the circuit from the LED through a resistor to what I ASSUME is a 2N6426 NPN Darlington transistor. This transistor has continuity from the collector to the emitter as measured IN CIRCUIT. I assume it is shorted. The other 4 Darlingtons in the circuit do not. I ordered transistors which will be here end of the week.

Can anyone tell me if the Darlington marked 6426 4 on the switching matrix schematic controls current flow to the inductor coil of RY4? I ask this because I ASSUME channel 3 is always on no matter where the channel switching is set based on the LED always being lit. RY4b connects channel three per the block diagram.

Other strange stuff

When channel 2 is selected the channel two LED comes on as expected but no sound from the channel until the master on channel 3 is turned up. At that point the channel 2 gain control works and sound is produced but a small movement of the gain control produces a huge volume increase. Also none of the channel 2 tone stack will work. There are other things I really do not understand about the 18 pin IC LM3914 like zero voltage at pin 5 turns on channel 2 and so does 1.3v...I measured the voltage needed to turn on a relay using a DC power supply at somewhere near 1.4v.

All this stuff is what happens when solid state engineers get involved in building tube amps...lol

Cheers,

Billy
Schematic link http://schematicheaven.net/boogieamps/boogie_nomad45.pdf