diytube.com

I have this magnavox amp and I am trying to use it for guitar. Tho it sounds pretty good as is for music, it sounded bad for guitar. I would like to try and improve it somehow. Do you guys have any suggestions?

The first thing I did was clip the negative feedback wire in such a way that I could easily add a switch later.This one thing has made it sound much better and now I am excited about the thought of stereo guitar effects! Next I would like to make the pentode a triode by connecting pins nine and seven to see how that sounds. One question, does that kind of global negative feedback reduce the output impedance? Triode strapping reduces output impedance because of the local feedback within the tube right? So would triode strapping + no negative feedback = pentode + negative feedback here?

Another thought was to use a different nine pin output tube, Maybe a 6s4, or even a 6fd7 to add a gain stage. So far it doesn't seem to need more gain and it is very loud as is but maybe it will need more gain in triode mode, I didn't try it yet. I'm trying to go for the most even order harmonic distortion possible and the least amount odd, hence the idea for the triode conversion. I appreciate any help thanks, Austin

Triode strapping the 6BQ5 effectively lowers its internal plate resistance (RP) to about 2000 ohms, from the pentode connection's aprox 38,000 ohms. IT also significantly reduces the tube's gain, which means less clean output power, but then that is not all bad.

That, in turn, will improve the damping factor, but the downside is that a triode strapped 6BQ5/EL84 will deliver only about 2 watts rms output, while a pentode is good for about 5 watts rms. Triode mode doesn't generally need the NFB, as much as the pentode mode would require for most applications.

The tonal quality will also be quite different, the triode in overdrive, will produce mostly even (2nd, 4th, etc) "octave" harmonics, and the pentode is more prone to the 3rd and 5th harmonics, which tend to be more dissonant. It is all a matter of personal taste and the music being played.

FWIW, the rather complex NFB loop in the amp schematics looks like it has some built in bass and treble boost, which would help make up for the shortcomings of the small OPTs and inexpensive speakers, to a point

For guitar use, this would need another gain stage up front, use a 12AX7 or 5751 with 1.5K RK and 100-120K plate load resistor, and a small 0.1 - 0.22 uF coupling cap. The added input stage should have a 1 meg grid resistor, and a 68K resistor in series with the "tip" from the 1/4 inch input jack - this is a very "standard" guitar amp input configuration. The volume control (and tone stack) would connect in between the first and second gain stages..

HTH

/ed B

Is a particular damping characteristic that important? One of the all-time great guitar amps used EL84s in pentode mode with no negative feedback (I'm thinking the Vox AC-30, here). Consider also that a common trick used to make solid-state guitar amps sound more tube-like is to increase the output impedance. It becomes a tonal choice: an amp with less damping lets more of the particular quirks of the speaker come through, which is a plus if you have speakers that behave in those conditions, and a minus if they don't.
Also, one of the advantages to pentodes is that by changing the load, you change the distortion spectrum. With a 5K:8 output transformer, hooking up a 4 ohm speaker will give you a 2.5k load which makes a lot more 2nd harmonic than 3rd, and with a 16 ohm speaker, you'll get a 10k load which will give a lot more 3rd harmonic than 2nd. If you find a good EL84 datasheet (the GE 6BQ5 one is nice, if I recall) it will give a chart of distortion vs. load for some voltage, which you can use as a reference.

For a guitar amp, the damping factor isn't very important, as the lowest frequency is about 82 Hz, and, a bit of distortion is a good thing,
especially for getting the tonal quality lead playing. An overly "clean" amp just wouldn't sound right, or even very good, (actually it would tend to sound flat and lifeless) for guitar applications.

DF is a more important consideration for clean, deep, bass, without the "flab" or "fartyness".

/ed B

Thanks for the info! Can I connect the screen to the plate directly or should I use a resistor to limit screen current? Would omitting the cathode bypass capacitor help reduce odd order distortion specifically or just all distortion in general? Thank you

I prefer to use a resistor value, between 100 and 1000 ohms to connect the screen to the plate in triode strapping. 470 ohms, 1 Watt is a good "standard" value, which I generally use (and because I have a "ton" of this value in my parts collection)

This is also a good value to use in an "ultralinear" connection, between the G2 and the SG tap on the OPT, if it has one available.

The actual screen current in this mode is usually between 1 to 2 mA, so the actual voltage drop would be small, no more than a volt with 470 ohms. This resistor also eliminates any possible spurious RF oscillations, and protects the screen better than a direct plate to screen conenction.

The 6BQ5s; cathode bypass cap should be in the circuit, else the power is going to be significantly reduced, and the lower few octaves' of frequency response will suffer without bypassing. In other words, it would sound kind of "nasty" without the caps in there

HTH

/ed B

EWBrown wrote:
The actual screen current in this mode is usually between 1 to 2 mA, so the actual voltage drop would be small, no more than a volt with 470 ohms.


/ed B

So may I use the 1/4 watt resistors I already have since its only 2 ma? thanks

So may I use the 1/4 watt resistors I already have since its only 2 ma? thanks

Quarter watter should be OK, as the actual dissipation through it would be just a couple of milliwatts. Or about 1% of the wattage rating.

I just got into the habit of using 1 or 2 W CC or MF resistors anywhere the DC or signal voltage across it could exceed a couple hundred volts, or anywhere near 350-500VDC B+. More of a matter of internal spacing, for higher voltage considerations, than for actual power being dissipated. .

/ed B