Problem with the high hum in a tube amplifier for a guitar

tube no input ac voltage hum

The small change can be expected from grid current in connection with 475K resistor or from JFET leakage current, and isn't critical, I think.

Carry on, good luck!

take away hum tube amps

Hi Frank,

My amp is still playing tricks with me! I checked the solder connections and found that the gate at J175 at V2b grid was not connected propperly! Soldered it and now my ch1 is loud and really good sounding! ..But at V2b anode I measure 255v DC on channel 1 but when switching to channel 2/3 I measure 63v DC. This is a huge voltage swing which totally kills the gain!

I thought could be a cold solder at 1uf/100ohm..however, I resoldered the 1uf/100ohm but i didn't help! When I measure the resistance between the 100ohm res and ground, I get 100ohm so guess it's connected propperly!

What else could it be?

Thanks

Thomas

Added after 14 minutes:

And! Just measured 6v DC at the grid! this should not be there, right! huess it must come from the J175 which I've substituted with 2n5460!

I'll try to see if I've got the pin-connections right..

Added after 4 hours 4 minutes:

The pin connection was wrong! ..I replaced the 2n5460 and now everything is working propperly except from ch1 which now have too much gain Think this is because of the 2n5460..

remove humming amplifier ground problems

Hello Thomas,

according to datasheet, neither J175 nor 2N5462 should have a high gate leakage current. This behaviour is clearly irregular. Most likely the FET had been damaged. JFET are not very sensitive, but high positive voltages above the 40 V breakdown voltage at the gate may cause permanent degradation and possibly higher leakage current. Can happen e. g. when soldering it with an iron without protective ground connection.

I think, the gate series resistor, enabled by the FET in channel 2/3 operation has the purpose to give signal compression or sustain. Shorting the JFET gate to ground bypasses the 475K resistor and disables this function, but would allow normal operation for the present.

Frank

killing the hum vacuum tube

If I measure at the 2n5460 gate get 0.6v DC at channel 1 and 15v DC at channel 2/3.

In channel 1 I can only turn the gain pot and master pot to about 1 out of 10, before it blows my head away and clipping the signal! Channel 2/3 are also very loud, so I think I might have damaged the 475k grid resistor.

I dont the FET is there to "change" the sound (such as sustain/compression) since the amp/power supply already has two rectifier tubes which produce alot of tube sag/compression..But not sure at all! It's realy the only part of the switching circuit that confuses me (also because of the symbols in the scheme since the J175 is a P-channel fet and I mean that the little arrow should follow the gate' direction and not the source' direction).

However, I'm not sure if the 0.6v at the gate is as should be or not..but this should explain how the FET is switching.

Thanks

Thomas

Added after 7 minutes:

btw..my solder iron is an old weller WECP-20 and do have a ground terminal..should I connect it to the chassis ground???

why amp humm

Hello Thomas,

yes, an open circuit 475k resistor would also be possible reason for getting an undefined grid voltage. But the grid current is generally negative, if a measurable amount exists at all, simply by collecting some electrons. So a positive grid voltage as you measured it and as would be necessary to pull down the anode voltage, can come from leakage currents only. I think, they can be expected from the FET mainly, specified maximum value is 5 nA however. Otherwise, other leakage sources, not viewable from the schematics, must exist.

I understand, that you measured 6 V at the grid, I assume with 10 Mohms multimeter or ocilloscope probe, so the leakage current would be at least 0.6 uA.
But there are other differential diagnostic means to check which part is actually faulty.

The 0.6 and 15 V at the FET gate is what I expect from the schematic, the circuit operation is completely understandable so far, only the purpose of the grid series resistor is in doubt. I still think, that it has to do with dynamic behaviour in distorted operation. You can see, that channel 2/3 has higher grid resistances than channel 1.

I think however, that the circuit details are empirical to a high degree. Even the original designer must not always know, how it's operating exactly. I've seen a collection of different Mesa Boogie schematics, kindly published at EDAboard, and none was exactly like the others. The guys are working in craftsman's manner apparently.

Regards,
Frank

P.S.: Weller WECP should be O.Ks as is, I never damaged any part by this tool.

tubeamphum

Hi Thomas,
Glad that you are making some progress with your amp now. Frank is doing a great job, he is and excellent engineer.

I think I will have a go at building that amp myself, as I need one for the electric guitar that I made.

Best regards,
Bob.

tube amp star grounding

Hi Bob!

The amp is just awesome!

The channel 2 (crunch) is extremely hard sounding and channel 3 (Lead) got a very smooth and still hard sounding distortion..besides that, the many voicing features makes the amp capable to produce the softer marshall kind of sound..I play in a metal band and used the marshall EL34 100/100 and JMP-1 rack series (which I used to love) but this new build is totally blowing it away!!!

It's the most powerfull sounding monster I've heard for a long time! I run it with 2xKT88 for a tight low end and 2xKT66 for nice mids and highs, all with individual bias pots! I play sevenstrings so the tight buttom end is a must! (Which marshall never was capable of).

I've pretty much solved all my issues with the amp now besides that the amp is clipping at V2a. Think it's because of a possible defect 475k grid res.

However, if you are planning to build this monster, be prepared to drill approx 100 holes in the chassis It got 27 pots and 7 switches and bla bla bla...

Still need to build the footswitch but it's almost done..

And Yes!! Frank is the man! I've posted this topic in three forums without getting any response..think this could be because no one has build a clone of this amp before and the missing parts of the switching circuit and 11-12 relays and 3 channels makes it hard to get an overview of the amp. Most respondents answer; try swap tubes or it could be a switching problem..ha ha..but frank realy did the job. I am totally new in amp building (..and electronics in general) and I realy learned a lot through this troubleshooting. It's so hard to get a realy good book! I've used www.aikenamps.com a lot and think he has got the best describtion of how these things works..however, I clearly miss some fundamentals in electronics but I am getting better

I still need to check the Fx-loop, but first the 3 channels must work propperly and stable.

Best regards

Thomas



Thomas

ac heater tube hum

Hi Thomas,
A bit off topic but probably OK now that you have got the amp about sorted.
It was interesting to hear your ideas on the different tubes and their effect on the sound colouration, that makes complete sense to me. I would be interested to hear if you think there is a noticible effect on the sound colour when using solid state diodes or thermionic diodes for the high voltage rectifiers as this circuit has the option of both and the makers do make a thing about it in the name of the kit. I have never done any comparative tests, so would like to know your views as you are an experienced musician. I would assume that there might be some effect as the forward resistance will be quite different but this effect is probably heavily modified or even almost removed by the high voltage DC smoothing components. I wonder if Frank has any thoughts on this one too?

I am certainly sold the design concept of this amp but will probably try a few minor mods when I build mine up.

Thanks for all the help points that you have sent.
Bob.

hum in tube amplifier

Hello Bob,

I didn't notice the rectifier as sound forming amplifier element before seeing the Mesa Boogie Dual-Rectifier. Thomas recently stated, that it matters regarding compression. The effect is basically understandable, but it iworks at full output power only, I think. So it may be difficult to use it at your home.

I think that the Dual-Rectifier is a very special design. It tries to offer any available amp variant as a switchable option. Very flexible - but requiring a lot of controls. But I must admit, that my practical knowledge with guitar amps is mainly limited to antique models as VOX AC30 or Marshall 19xx.

Regards,
Frank

tube amps ground loops

Hi Bob/Frank

I haven't got the chance to test the different sound modifications yet. I tested the amp yesterday on the big 4x12" cabinets and it realy sounds awesome. But it's getting way too hot. After 1 hour the amp was loosing it's power because of thermal stress, I think. The amp was standing upside down so the heat from the output tubes was going directly towards the chassis, I could not hold my hands at the rear side of the chassis!
This limited the test, but I will install some fans cooling directly at the power tubes.

I compared the noise level with a carvin legacy and the marsahll stack and in comparison the amp is very quiet so I am fully satisfied with the result!

I've desoldered the J175 (2n5460) and measured 450kohm across the 475k grid res at V2a, so installed a 22kohm res in series. This cured the clipping at V2a for channel 1. However, the basicaly don't have any headroom at all! If I turn the master 0.5 - 1.5 out of 10 (at any channel) the amp is LOUD. The marshall amp will have to be at 6 - 7 out of 10 to get that loud. As I wrote, I've desoldered the J175 (2n5460) at V2a grids which function might have something to do with this issue.

So guess I still have some issues to solve..any suggestion are welcome

Bob, As Frank wrote the amp is very very flexible (notice for example the switchable power supply. besides the rectifier selcetion it also had the Spongy/Bold switch at the primary which as I understand will give the amp more or less voltage/gain!)

what changes would you make to the schematic?

Added after 10 minutes:

One little thing! We are recording some new songs right now and offcourse we recorded the amp. This was how we noticed the power drop since the record in level bacame lower. However, my friend deleted (unfortunately) the recordings and we had to switch to the marshall stack. This was here the fun thing took part! We all realy noticed how different the two amps sounds and we decided to put the recordings on hold untill my new build is finished Offcourse this don't mean that the marshall is bad sounding, it's just an expression of that we're all damn tired of the marshall sound!


Thomas

humm problems with tube amp

Hello Thomas,

I still think that the V2a FET is defective (showing abnormal leakage current). The 450k to 470k difference shouldn't be noticable at all, it's within a usual 5% tolerance, actually.

Regarding gain control operation, I would expect logarithmic potentiometers for gain and linear for frequency controls. If you installed linear type, this may be the reason for different behaviour compared to other amps. I didn't find a hint in the schematics, you may have an opportunity to check with a series Mesa Boogie or ask a guy involved in amp repair.

Log ("B") or lin ("A") is mostly printed near the resistance specification at the part. The type can be easily checked with a multimeter when setting the control to midrange and comparing the partial resistances.

Frank

tube amp high frequency hum

Hi Frank!

The gain pot's don't have influence of the output level (not much). The amp is also available in a 2 channel version with basically the same preamp and same values. I've seen others who have build the 2ch version with succes. However, the 2ch version has a different switching circuit (LDR based). I use linear pots for everything except from the precense/middel/treble controls. The EQ shapes the signal but don't realy affect the the output level that much. It's a classic V-shape EQ.

I'll try to install a new 2n5460 across the 475k to see if it makes any difference, because I wasn't notice this small headroom when the JFET was installed. Don't understand it's function at all, besides that it could be limiting the signal with the leaking perhaps?
Thanks

Thomas

amplifier hum ground

Hi Thomas,
I agree with Frank, the CH1 (1MΩ), CH2, & CH3 (250kΩ) and the three 1MΩ master gain pots in the preamp should certainly be logarithmic because the human hearing response is logarithmic, these pots should all be able to kill the signal dead because they are ground referenced. I feel that there may be other gain pots in the amp that should also be logarithmic too. If these pots do not have much effect on the signal loudness, you are almost certainly missing a common ground connection between two parts of your amp, this would explain a lot of the problems you have encountered. Your amp is running at almost full volume all the time!

The 475k resistor value is not out of the manufactures design spec either as Frank points out, it is almost certain to be the FET leakage problem again that makes the 475k resistor look like a much lower value resistor than it should be, remember if the FET when turned off reads 1000k it will look like it is just under 320k to the circuit. Remember that a junction FET behaves like a voltage controlled resistor, it needs to have an off resitance of probably 10MΩ before it really looks to the circuit to not be there.

Best regards,
Bob.

how to improve tube amp hum

Hi Bob!

Thanks! I'll try to check the ground connections!

Best Regards

Thomas

Added after 49 minutes:

Btw..Where would you suggest to ground the pots? This is how I've grounded the preamp:

Ch1:
Gain and Mid at V1a
Master and Presence at V1b/V2a

Ch2/3
Gain at V1a
Mid, Presence and Master at V2b/V3

All Jacks are insulated from chassis and Output transformer/ouput jack (ring) is grounded at V2b/V3

Thanks

Thomas

amp hum ground

Hi Thomas
Sorry about the delay in getting back to you. Yes the ground points that you mention should be OK but are all the separate ground points all linked back together at the power supply common ground point? Avoid 'daisy-chaining' at all costs, this is where the grounds are all linked together but from one module to the next and then to the PSU they should all have their own wires all the way back to the common point, although on a PCB layout this is not always easy or obvious. Make sure that the ground wires/PCB tracks are thick enough too, especially the ones for the power amp return. Every ground symbol on your circuit should be linked together for all modules. You shold be able to link an ohmmeter to any ground point and it should read less than 1Ω to any other ground point on your circuit on any module. This is important, you must have got most of them OK or the circuit would not work at all but it is easy to miss one off (we have all done things like that at some time, even experienced engineers get it wrong sometimes, if you are distracted while wiring something up you think you have made all the connections when actually you have missed one!), which will cause the sort of effect that you are seeing.

Best regards
Bob.

grid resistor hum

Good Morning Bob/Frank

I think my PCB ground traces at the EQ's are too tiny/thin..I think I'll try to wire them PTP.

Btw, I've found this interesting article about grounding! It realy makes it all much more clear to me.

The C/D/E voltage capacitors are all laying very close to their tubes (where they are being used) but only whit the positive connection and not the negative (ground) connection. On my PCB I can see that V1/V2/V3 shares the same return path to the 'big' ground plane and V4/V5 has another return path. I think I'll try to make some 'loops' back to the tubes individual power supply.

Thanks

Thomas

heater cathode hum amp fix center tap heater

Hi Thomas,
Yes that article contains a lot of useful stuff, the last paragraph speaks volumes.

This grounding is the biggest secret there is to successful analogue electronics. There is a hell of a lot more to it than most electronics buffs think in their early experiences. By going through what you are now going through now you will learn so much that will benefit you in all your future projects. I well remember being shocked at how many points an old experienced analogue engineer that I used to work with pointed out in one of my very early designs, he was right of course! You keep on learning all the time in this game, I have been doing it over 30 years and I am still learning new tricks from every project I do!
Bob

how to solve humming sound from tube amp

Hello,

Circuit paths not shown in the schematic is an important and basic point in the paper. The Mesa Boogie schematic is a good illustration. It is intended as a service and repair document, not an application note how to make a good tube amp. Thus it doesn't show the ground wiring.

The Mega Boogie designers may have internal docúments, that show the ground net topology, probably that don't have, they simply know about, as Bob and I would know.

If manual wiring is involved, explicite assembly instructions should exist at the factory. I say should, cause I know of small factories working in craftsmans manner, where assembly instructions exist mainly as oral tradition. A particular company is manufacturing instruments, I designed 15 or 20 years ago. Sometimes I'm wondering what the guys are doing there, when I drop in. But they got ISO 9000 certification and some assembly instructions since then. Digital photos of reference instruments play an important role.

Regards,
Frank

my tube amp is humming

Hello Frank/Bob

I'm well familiar whit the oral tradition and ISO 9000 I am newly graduated master of science specialized in production engineering..so the oral traditions and the hidden knowledge are something we are trying to avoid and get on paper

However, I'll try to review my groundpoints with the article in mind and let you know the results.

Bob! the fun thing is that know, when I'm almost through with this project, I am actually considering building it all over again, because of the the new learnings/findings!

Thomas

Added after 2 hours 23 minutes:

Hi,

I've resoldered my offboard wirings but it did nothing to the headroom. All ground connections are good but all three channels are way too loud. Can I somehow measure where the problem is? I measured the resistance between the wiper and ground of each master pot. They all show a max resistance of 3kohm when their channel is selected.

Thanks

Thomas

Added after 44 minutes:

Could I solder a resistor between the treble and master pots and within which range would you suggest?

low-level amp hum

Hello Thomas,

as said, logarithmic pots for volume would most likely remove the problem. A log potentiometer gives typically 5% of maximum gain at mid position. The problem is, that the 1M load of the master volume pots is most likely required by the preceeding circuit. Without assembling new pots, I would try a shunt resistor between master volume wiper and ground, e. g. 100 k. But it will give less basses at maximum volume setting.

Regards,

Frank