Maker Pro
Maker Pro

Valve (tube) amplifier schematic help

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
An inefficient way is to put a lot of idle current thru the fet and reduce the source resistor to a very low value. This will flatten the battery quickly though, and need a big heatsink.

I'm interested in this, can you give me more specifics?

Battery life is not an issue. This will be powered off 120VAC wall.

This secondary project is already accounting for the use of efficient heatsinking as well as a fan in the case to dissipate heat.

That is, unless the inefficiency comes at the cost of audio performance.
 
Last edited:

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
Using a large current through the mosfet will not give great hifi, since the linearity of the device comes into play. This called a class A power amp and requires a bit of attention to the DC bias to make it work. I can't be bothered with the specifics of this, it is more trouble than it is worth. Just use an LM386 power amp chip after a preamp triode, that's easiest.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
circuit2.jpg


After taking a very detailed look at the datasheet, it says pins 1 and 8 can be jumpered with a capacitor together to enhance gain. I, in the diagram, placed a trimming varicap at pins 1 and 8 to allow for adjustable gain. I paralleled it to the right channel to make the effect even.

Does this work?
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
I can't be bothered with the specifics of this, it is more trouble than it is worth. Just use an LM386 power amp chip after a preamp triode, that's easiest.

Love the attitude. You're going to work out well here. :D

Welcome!
Chris
 

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
First, you don't need 2 triodes. If you look at each one, you'll notice you are only using 1 triode in each tube. You can just use i tube, since it is a dual triode. You should not share the capacitor betweens pins 8 and 1. Since thge datasheet specifies about 10uF, you can't use a trimmer. Just use 2 caps. Pin 7 should not be grounded but needs a cap between it and ground.

The caps at the outputs of the LM386 are not right. Sries caps are of no use here. There should be no cap to ground unless you are using the zobel network as suggested. Then you must have a series resistor too. This will stop ultrasonic oscillation.

You should go through and calculate the total voltage gain and see if it meets your requirement. Also, the stereo pot has the hot and wiper swapped.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
The use of two tubes is an aesthetic design choice. I don't need the LEDs coupled on to the B rail either.

Is there any purpose to coupling the output at one anode to the grid of the next for two stages of amplification?

Respond if you want, but I may or may not respond for a few days. It seems I'm doing something that Einstein once described as "hunting for birds in the dark in an area with very few birds".

I, this entire time, have been trying to create a circuit that I could put in an enclosure and call my own. That's the reason I haven't gone out, bought a premade kit and just assembled and soldered it. Well, seems like I've hit the point where the plans carry little or no feasibility and everything I try to draw up or do differently end up creating another error, so I'm going to take a couple days to reevaluate whether I want to continue trying my hand with these projects and amplification ideas.
 

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
The use of two tubes is an aesthetic design choice. I don't need the LEDs coupled on to the B rail either.

Is there any purpose to coupling the output at one anode to the grid of the next for two stages of amplification?

Respond if you want, but I may or may not respond for a few days. It seems I'm doing something that Einstein once described as "hunting for birds in the dark in an area with very few birds".

I, this entire time, have been trying to create a circuit that I could put in an enclosure and call my own. That's the reason I haven't gone out, bought a premade kit and just assembled and soldered it. Well, seems like I've hit the point where the plans carry little or no feasibility and everything I try to draw up or do differently end up creating another error, so I'm going to take a couple days to reevaluate whether I want to continue trying my hand with these projects and amplification ideas.

There certainly is a purpose. It is to create sufficient gain to amplify small signals. With 2 stages, it allows a volume control to be used toward the lower end of its range, it is how we are used to operating audio equipment. If you would like to use 2 triodes, do this, just make the redundant triode in each tube a second stage. Then you can amplify very small signals.

Random fact: many big Marshall guitar amps only need 0.5mV to reach full power at the output. This gives an idea of the kind of gain in such a thing.

It seems to me you are attempting some challenging design apects for a beginner. That is absolutely fine, but you need to be prepared to build something that does not work at first, then debug it, That is the best way to learn. I appreciate you are trying to make original stuff and that is worthy. It is easiest with this approach to develop a working project into something unique after getting it working first. Then add your own original ideas.

To make a guitar amp from scratch, check out the AX84 project. These things sound great and are basic. They are not kits but there are schematics and different versions on the website. There are helpful people there too to help you get it going. You would have to put in a LOT of effort to come up with something that sounds as good. I made one called an SEL for my son, I just used a different output tube to increase power. You can do endless tweeking too for tayloring your sound, as far as component values are concerned.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
If you would rather use valves in the output stage of your headphone amplifier, Google
"class A" "headphone amplifier" valve tube "output transformer"
The first five hits are all interesting.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
Well, I've spent the last few days cutting and etching PCBs.

Figure it's time for me to start asking less questions and more building and finding out the hard way, if you will.

Regarding the schematic Chris posted earlier (power supply), with two 100uF and one 100nF caps for ripple filtering, can the 100nF be omitted safely?
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Regarding the schematic Chris posted earlier (power supply), with two 100uF and one 100nF caps for ripple filtering, can the 100nF be omitted safely?

The 100nF cap is a decoupling cap that should be placed close to the supply side of the plate resistors. There's no good reason to omit it. They're dirt cheap!

Chris
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
The 100nF cap is a decoupling cap that should be placed close to the supply side of the plate resistors. There's no good reason to omit it. They're dirt cheap!

Chris

I was just having a hard time finding them on Mouser in that specific capacitance, that's why I was asking if it was alright to omit it.

Chris, do you mind if I PM you to ask you some more questions about how impedance works?

From an amplifier design standpoint, I understand the absolute basics of how impedance functions in a circuit, but people who design amplifiers don't get by on just the basics, and I'd like to eventually be able to design my own amplifiers on more than just educated guesswork.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Having a hard time finding 100 nF capacitors on Mouser? That's like having a hard time finding Cheddar in a cheese shop! "It's the single most popular capacitance value on earth!" (to paraphrase Monty Python).

It's also referred to as 0.1 uF, or just "point one" :)
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
Haha. Nevermind that statement, I found them, but they had to be backordered. I was sitting to myself thinking "you have 5,200 of everything else in stock!"

and for whatever reason, apparently American convention is to make believe the nano-farad convention doesn't exist. I can't honestly be bothered to wonder why.

then again, I think the idea of calling a thermionic emitter a valve illustrates its purpose a lot better than a "tube", although "tube amp" to me sounds better than "valve amp".

Should be a show - "American Terminology Idiosyncrasies" haha

But for example, why is it I can't just put a bunch of resistors on the tail end of the circuit to lower its impedance?

Would that work (at the cost of sending power through the floor)?
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Haha. Nevermind that statement, I found them, but they had to be backordered. I was sitting to myself thinking "you have 5,200 of everything else in stock!"

and for whatever reason, apparently American convention is to make believe the nano-farad convention doesn't exist. I can't honestly be bothered to wonder why.

<snip>

But for example, why is it I can't just put a bunch of resistors on the tail end of the circuit to lower its impedance?

Would that work (at the cost of sending power through the floor)?

You're correct on that count. We don't use nanofarads in the US. In fact my schematics wouldn't display it as such except for the fact that when I enter .1uF Tina automatically converts it to 100nF. How the hell can a supplier be out of stock on a .1uF cap????

Regarding just hanging a bunch of resistors on the output to lower the impedance...

We're quite deep into this thread so I don't feel like searching back though all the posts to find it but I thought I covered that with a spice simulation earlier. Maybe not. In short, no you can't. All that will do is swamp out your available signal. If you were driving a PA with an input impedance of about 1K you could probably drive it with a Cathode Follower or use a JFET Source Follower. Neither would be sufficient to drive a speaker though.

Chris
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
You're right, you did run a SPICE simulation on it and show it to me, but my questions are in the realm of:

What makes devices high-impedance?

Are there practical ways of lowering this impedance without having to use specialized means (earlier in the thread I was told to use a LM386 op amp)?

I understand that a transformer-based device called a balun is used in settings to lower the input impedance. Would this hold for the output as well?

Also, an unrelated question, but does the notion of plate-loading resistors apply to the collectors of transistors as well?
 
Last edited:

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Vacuum Tubes are inherently Hi impedance devices. As Kris explained to you earlier, they can be viewed as an object who's resistance between the Cathode and the Plate is varied by voltage changes between Cathode and the Grid. Since they operate with relatively high voltage and low current they are high impedance devices.

BJT's, are current controlled devices that work with much lower voltages and conversely with higher currents. They are much lower impedance devices than vacuum tubes.

The value of the plate load resistor is related to the conductance of the tube but Kris covered this with you already. I see no point in rehashing it.

As I've stated repeatedly, our tutorial section covers all of this in great detail. Please visit it.

Your amplifier was not intended to be a power amplifier. For this reason a transformer is not an option for you.

Chris
 
Last edited:

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
Vacuum Tubes are inherently Hi impedance devices. As Kris explained to you earlier, they can be viewed as an object who's resistance between the Cathode and the Plate is varied by voltage changes between Cathode and the Grid. Since they operate with relatively high voltage and low current they are high impedance devices.

BJT's, are current controlled devices that work with much lower voltages and conversely with higher currents. They are much lower impedance devices than vacuum tubes.
That makes more sense.

The value of the plate load resistor is related to the conductance of the tube but Kris covered this with you already. I see no point in rehashing it.
I understand that, and that was not what I was asking.

I was asking if plate-load resistors have an analog in transistor-mediated circuits.

As I've stated repeatedly, our tutorial section covers all of this in great detail. Please visit it.
I've lived there, and I've been learning from not only it, but other sites, but the key to understanding is experimentation, and I don't have a wide selection of components to work with, so most of my questions are fulfilling that role.

Your amplifier was not intended to be a power amplifier. For this reason a transformer is not an option for you.

I wasn't asking for the purpose of trying to use a transformer on this circuit, as its amplification factor is too low. I was asking about future circuits and designs.

To tell you the truth, this pre-amp stage has almost killed me and not just in terms of stress.

I will tell you that I had this pre-amp working perfectly for 5 minutes. Came out great, good valve flavor, but I turned the unit off for a minute to go get a guitar and run through it, no go. Haven't gotten it working since.

Got the oscilloscope, was going to take readings at 100, 1000, and 10kHz to try to get an idea of what's going on with the thing, but the thing was slow to turn on, so I checked the tubes in the back. Checked out, put cover back on, but hand hit the on switch on accident (should have unplugged - nube mistake) and I juiced my finger with no less than 350V off the cathode ray tube. I have photos of the two-contact pinout of the tube burned into my finger if you don't believe me.

"I didn't have an appropriate resistor, so I used my finger instead"

Came back next day, rewired circuit on breadboard, giving everything extra space to hopefully compensate for stray capacitance, tested, but apparently I wired it wrong or lost track of a wire, as for whatever reason 12VAC was flowing out of the input line to guitar. I realized this the hard way when I held the metal jack with one finger and touched the strings.

I already know I'm going to get a lot of flak for these nube mistakes, but right now I'm trying to learn from my mistakes (what I just posted here) and less about doing insane modifications to a pre-amp valve to hopefully get it to work.

On the bright side I did go out and invest in a digital multimeter (not the ghetto analog one you've probably seen in my photos) that has increased range, frequency, diode and capacitance testing capabilities.

Eli
(solidus)
 
Top