Madamp M15MK1 build help....

Tube or solid-state, this section goes to eleven!
User avatar
bmxguitarsbmx
Cap Cooler
Information
Posts: 510
Joined: 25 Mar 2010, 21:15
Has thanked: 700 times
Been thanked: 216 times

Post by bmxguitarsbmx »

Alex's idea is a good one. For a quick solution, swap your tubes around until you get one that doesn't conduct any grid current. Grid current can start to flow at low plate voltages, but your design has plate voltages that should be ok for a majority of tubes out there.

User avatar
ryanuk
Solder Soldier
Information
Posts: 211
Joined: 29 May 2008, 11:18
Has thanked: 29 times
Been thanked: 28 times

Post by ryanuk »

Hey guys - bumping this thread.
I shelved this amp for a while as I moved and kinda forgot about it. Dusted it off the other day. Want to use this amp (Madamp M15MK1) as an opportunity to further my knowledge of tube amps with a view to working up to building my own.

Anyhoo, the amp still works but still has the issues of grid current, and my guitar volume is scratchy as a result. I note the suggestion of a coupling cap at the input.
However, looking at the Marshall 18w schematic, I can see they've utilised the design in the amp I bought. But the marshall plate voltages are a lot higher. With 280v before the 100k plate resistor. The Madamp has 250v at the same point. So the lower voltage on the plate (<150v) could be causing the grid current.

I've done some reading; I consider 2 approaches...

(1) could I raise the plate voltage by tailoring the plate resistor?
Would I need to raise or lower it? Schematic shows a voltage drop across the 100K plate resistor of 117v - so current of 1.17ma.
I'd read that 0.8ma and 150v was about right for the Marshall 18w so if I used a ~130K it would put me in the ball park?

(2) Or perhaps lowering the B+ node resistor (R16) to increase the B+ and ultimately plate voltage.

Can you guys comments on the implications of these suggestions?
Schematics earlier in this thread.

User avatar
bmxguitarsbmx
Cap Cooler
Information
Posts: 510
Joined: 25 Mar 2010, 21:15
Has thanked: 700 times
Been thanked: 216 times

Post by bmxguitarsbmx »

I'd stick the cap in. That will effect the tone the least of any solution. Raising the B+ 30V will change the tone a lot.

User avatar
ryanuk
Solder Soldier
Information
Posts: 211
Joined: 29 May 2008, 11:18
Has thanked: 29 times
Been thanked: 28 times

Post by ryanuk »

bmxguitarsbmx wrote: 24 Nov 2021, 03:44 I'd stick the cap in. That will effect the tone the least of any solution. Raising the B+ 30V will change the tone a lot.
Thanks @bmxguitarsbmx.
I will absolutely try that.

But, for the purposes of learning/experimenting, I was keen to understand how to adjust the node resistor/plate resistors for adjusting the voltage on the plate.

My confusion largely comes from looking at the schematics, and using ohms law to work try and work it out. I can see the approx. voltage drops at each point, and the R values, to get an appreciation of the current in the circuit at each stage.

I can see how I could adjust the Rs to get a different voltage drop, but then this would change the current at each stage?
Would yo select an R value which gives the desire drop as well as keeping the current requirements broadly as they are now?

User avatar
ryanuk
Solder Soldier
Information
Posts: 211
Joined: 29 May 2008, 11:18
Has thanked: 29 times
Been thanked: 28 times

Post by ryanuk »

I guess, asking my question a slightly different way...
Annotation 2021-11-24 145717.jpg
Annotation 2021-11-24 145717.jpg (16.79 KiB) Viewed 1711 times
This is the preamp Ive built from the kit. Voltage at the 100K load resistor is 250V - voltage at the plate is 133V. Voltage at the cathode is 0.9V.

Annotation 2021-11-24 145718.jpg
Annotation 2021-11-24 145718.jpg (16.79 KiB) Viewed 1711 times
This is taken from a Laney schematic. Voltage at the 100K load resistor is 200V - voltage at the plate is 140V. Voltage at the cathode is 0.9V.

How does the latter example start with a lower B+, but end in approx. the same plate and cathode voltage?

User avatar
bmxguitarsbmx
Cap Cooler
Information
Posts: 510
Joined: 25 Mar 2010, 21:15
Has thanked: 700 times
Been thanked: 216 times

Post by bmxguitarsbmx »

Individual 12ax7's will all bias slightly differently. Especially in this range where some 12ax7's will conduct grid current, which then causes more current to flow (as grid current makes the grid more positive in respect to the cathode, thus drawing more current).

Electrons from the cathode are drawn through the control grid to the more positive plate. As the plate voltage becomes lower, some of those electrons are able to accumulate on the grid and flow through the connected circuitry.

I have several amps that draw grid current at around this 140v plate volts area. I swap my tubes around to find the one with the least grid current. Grid current will cause pooping on relays too btw. That area can sound cool though, so I tend to ignore it. EH tubes seem to be able to get the lowest plate voltages without grid current and several other people I have chatted with agree with this observation.

User avatar
Manfred
Tube Twister
Information
Posts: 1936
Joined: 04 Apr 2009, 23:42
Has thanked: 1668 times
Been thanked: 1344 times

Post by Manfred »

If a valve in the idle mode draws such a noticeable grid current that the bias point is shifted, then the valve bulb is leaking and no longer has a perfect vacuum,
the air infiltrates and the tube is then defective.
The difference of the measured voltages comes from the current componet values and the operating voltage value at the measurement.
These values are different from amplifier to amplifier due to tolerances.
MADAMP.png
For example, Rk tolerance is -5% gives 780R.
MADAMP Rk Toleranz.png

User avatar
bmxguitarsbmx
Cap Cooler
Information
Posts: 510
Joined: 25 Mar 2010, 21:15
Has thanked: 700 times
Been thanked: 216 times

Post by bmxguitarsbmx »

Many tubes conduct grid current without being defective. Nothing wrong with the tube at all.

Those load lines are drawn with the grid connected to a low impedance bias supply that keeps the grid voltage stable. They will look different with high impedance's on the grid at low plate voltages.

User avatar
Manfred
Tube Twister
Information
Posts: 1936
Joined: 04 Apr 2009, 23:42
Has thanked: 1668 times
Been thanked: 1344 times

Post by Manfred »

bmxguitarsbmx wrote: 25 Nov 2021, 19:44 Many tubes conduct grid current without being defective. Nothing wrong with the tube at all.

Those load lines are drawn with the grid connected to a low impedance bias supply that keeps the grid voltage stable. They will look different with high impedance's on the grid at low plate voltages.
Yes, you are right that a grid current flows, but it is usually less than 0.1 microamperes for the 12AX7 (ECC83).
That's why I wrote about a noticeable grid current in the case of a valve with bad vacuum, that the operating point would be noticeably shifted.
With a grid resistor with a value of 1 megohm, there would indeed be a voltage drop of 0.1 volts.
Theoretically, one would have to increase the magnitude of the grid voltage values in the characteristic curve by this 0.1 volt,
but I think that one can neglect this due to the tube tolerances.
That's why the operating point setting with grid current and a high grid resistance value with the cathode to ground is an uncertain thing.

User avatar
bmxguitarsbmx
Cap Cooler
Information
Posts: 510
Joined: 25 Mar 2010, 21:15
Has thanked: 700 times
Been thanked: 216 times

Post by bmxguitarsbmx »

You might be right Manfred. This might be a bad tube. I'm not sure what a normal range of plate voltage variances would be for normal manufacturing differences.

User avatar
Manfred
Tube Twister
Information
Posts: 1936
Joined: 04 Apr 2009, 23:42
Has thanked: 1668 times
Been thanked: 1344 times

Post by Manfred »

bmxguitarsbmx wrote: 01 Dec 2021, 01:22 You might be right Manfred. This might be a bad tube. I'm not sure what a normal range of plate voltage variances would be for normal manufacturing differences.
The manufacturing tolerances of the tubes should usually be +/- 5%, at least for the good brands.

User avatar
phatt
Transistor Tuner
Information
Posts: 1224
Joined: 04 Aug 2010, 05:17
Location: Morayfield SE Qld AU
Has thanked: 8 times
Been thanked: 343 times

Post by phatt »

I'm guessing here, but as this amp has both clean and crunch grids tied then maybe that could cause issues. :scratch:
I would lift one of those 33k grid resistors on one or the other input grids and see if it helps.
Phil.

User avatar
bmxguitarsbmx
Cap Cooler
Information
Posts: 510
Joined: 25 Mar 2010, 21:15
Has thanked: 700 times
Been thanked: 216 times

Post by bmxguitarsbmx »

Good call Alex. Both grids will likely be conducting grid current. I consider anything less than 140V plate voltage something you might keep an eye on.

Install the cap as a DC blocker. The grid will still need to see a path to ground, of course.

User avatar
Manfred
Tube Twister
Information
Posts: 1936
Joined: 04 Apr 2009, 23:42
Has thanked: 1668 times
Been thanked: 1344 times

Post by Manfred »

Edit: Hi alexradium, sorry I got your name mixed up, i meant Ryanuk

Hi Ryanuk.,
I come back to your very first post and the actual problems.
You wrote there:
However, when I plug in, I get a horrible distorted gated sound (with a low level hum in the background).
Did I understand correctly, the preamp has this problem but the poweramp works if you feed the signal directly into its input?
Does the problem occur with both channels?
Is it dependent on the setting of the volume controls?
I guess the problem can not come the mains voltage deviations in the range of 220-240 volts.
The gating comes from the so-called blocking distortion.
This means that at a high signal voltage, a grid current flows at the positive half-wave, and the coupling capacitor is charged, and this voltage is then of negative polarity, which is added to the bias voltage at the grid at the negative half-wave,
thus blocking the triode. If now the signal voltage goes back so that d it takes a noticeable time until the capacitor is discharged again and the amount of charge accumulated between grid and cathode is dissipated and the tube can again amplify the signal properly. This delay is clearly audible.
The distortion is the so-called grid current or warm distortion. In amplifier stages where this is desired, a higher resistor is used to limit the grid current and thus the charging of the coupling capacitor, the delay time then becomes so small that it is no longer audible.
This grid current limiting resistor must be chosen carefully, because together with the input capacitance of the tube stage it forms a low pass and thus has a noticeable influence on the frequency response.
Have a look at this site:
Which pickups are installed in your guitar?

User avatar
Synthi
Information
Posts: 2
Joined: 29 Dec 2009, 17:04
Has thanked: 3 times
Been thanked: 1 time

Post by Synthi »

I fixed an M15MK1 by replacing the shielded wiring to the output tubes by unshielded ones...

Post Reply