Valve (tube) amplifier schematic help

[Solidus]

I didn't say anything about "automatic". I think we've reached a point in this thread that you should be pointed here..

https://www.electronicspoint.com/electronics-tutorials-f101.html

Doing otherwise could find this thread active in 2020!

Chris

I didn't mean automatic in the sense that it does it on its own accord, I meant automatic as opposed to a physical finger hitting a physical switch

And I don't mean to be overbearing, I just find that talking to an actual person makes a lot more sense than reading about it.

 

[CDRIVE]

I didn't mean automatic in the sense that it does it on its own accord, I meant automatic as opposed to a physical finger hitting a physical switch

Think of it more like a solid state replacement for an electromechanical relay.

And I don't mean to be overbearing, I just find that talking to an actual person makes a lot more sense than reading about it.

This is all well and good but the study of electronics begins with basic electron theory and a solid understanding of Ohms Law. Everything else is built upon these.

Besides, this is the 'Electronic Projects' section, not the Homework section.

Chris

 

[KrisBlueNZ]

This is true. Amplification could get so intense minute vibrations caused by air currents and static would go through the channel as well.

Yes, and valves themselves are microphonic. Vibrations from the speaker could easily be picked up by the early valves in the chain and cause feedback. Of course, some guitarists LIKE that, but it is impossible to control the frequency of the feedback.

Kris, you said you work with these things, what are the practical considerations of each?

JFETs vs. bipolar transistors... The main differences are 1. JFETs, like valves, are voltage-controlled and have a very high input impedance, while transistors are current-controlled and have a much lower input impedance - at least, when operated in the common emitter configuration, and 2. JFETs need a negative bias on the control electrode (the gate) (like valves do) (relative to the source), while bipolar transistors need a positive bias and some positive current flow into the control electrode (the base).

You two know not what you have done you have not only increased my confidence with this stuff a thousand-fold, but you have given me endless curiosity about it!

That's great! I agree with Chris. I think you should invest in some teach-yourself-electronics books. There are literally thousands of possibilities on amazon.com. There is a lot for you to learn, and it's a good idea to have it presented in a structured way.

The things that helped me most with learning were being able to visualise what was happening in the circuit (so, get a multimeter, and as soon as you can afford one, an oscilloscope), hands-on experimentation (so get a breadboard and a variety of components), and seeing how real products are made (so ask your family and friends to give you anything electronic when it breaks, then take it apart, try to figure out how it works, google the part numbers you find, and keep any parts that look interesting), and electronic kitsets.

If you get stuck on anything specific, there are many experienced and helpful folks here who are happy to answer specific questions.

Oh, and try to find an electronics club in your region. I WISH I'd had people who would look at my circuits and explain things to me. A club is hands down the best way to get this.

 

[CDRIVE]

Yes, and valves themselves are microphonic. Vibrations from the speaker could easily be picked up by the early valves in the chain and cause feedback.

This fact alone makes it nearly miraculous that the WW2 'Proximity Fuse' actually worked! The guys that designed and built the prototypes were mostly Hams of the era and they (in my opinion) were geniuses!

Chris

 

[Solidus]

Alright, after a few days' of doing my homework on the tutorial section, I'm back with more newbish questions

The amplifier is coming along well, just have to navigate the maze of wires from 3 different pots, two different phono lines, and 4 different switches. Tested everything for continuity, so far everything is solid.

I have to thank you two, you have created an insatiable interest in me for this stuff. I've spent the past few days designing endless circuits with valves and transistors, things that probably don't work the way I think they will but with time, I hope to be in your position one day, lecturing some newbie on obvious mistakes!

Right now I'm still trying to visualize the transistor as a triode valve minus the heater, it helps my confidence a lot when the object of consideration is a clear glass cylinder that can be studied and not a black square of mysteries and black magic (I don't really understand what goes on in there).

I'm going off to my first year of college in about two weeks so it shouldn't be too hard to pick up on some second- or third-hand electronics texts and maybe even a friend who can sit down with me on the stuff.

Also, I have a basic multimeter and old-school valve-driven oscilloscope, so with a little bit of experience, I should have the basic tools to do what I want to do.

While I'm in college and when I have a bit more confidence with my circuits I will end up rebuilding this amplifier anew, but with multiple valves and a completely homebrew circuit. Purchasing power supplies makes it easy on me experience wise, but doesn't allow me to have the satisfaction of a circuit 100% made by me.

Would either of you mind if I PM'd you to work on the electronics stuff as I work through it? The hardest part about teaching myself, I've found, is the hesitation in not wanting to start experimenting for fear of doing something wrong and having later things built on incorrect knowledge.

 

[KrisBlueNZ]

That's cool that you want to get into electronics. It's a really interesting subject! But of course I'm biased

It's great that you have a multimeter and an oscilloscope. You can do a surprising amount with an old analogue scope, even something with only 5 MHz bandwidth and a single trace. That's what I started with!

Re the difference between valves and transistors, I don't see how the fact that valves have transparent cases helps you understand them more than transistors, but I think the simpler principle of operation (electrons repelled by the negative voltage on the grid can't get to the anode) makes them easier to understand than bipolar transistors, which use doping and two types of current carriers.

I'm happy for you to PM me in the future. I think it's cool that you're so interested!

Good luck with your studies and learning about electronics

 

[Solidus]

You're probably right. The idea of valve operation sits better with me rather than N and P doping in a silicon wafer.

I just went out and bought a small breadboard and 3 different transistors (1 PNP, NPN and MOSFET), so the experimenting will begin!

I'm also working on wiring the mains supply for the amp. Should I route every grounding position on this circuit to the mains grounding line?

 

[KrisBlueNZ]

Should I route every grounding position on this circuit to the mains grounding line?

Normally you should ground circuits for safety, but that can cause earth loops (see http://en.wikipedia.org/wiki/Ground_loop_(electricity) ). In this case your power supply is double insulated so there shouldn't be a safety issue, and it will be grounded through the amp whenever it's plugged into it (assuming the amp is grounded, which it should be), so you shouldn't ground it directly to mains earth independently.

I suggest you have a look at that Wikipedia page and make your own judgement on whether there will be a problem and how to deal with it.

 

[CDRIVE]

I'm happy with your excitement and desire to dive in. No problem with PMs but I'll ask you to include the link to your new topic. This is because EP is an educational forum, PMs benefit only the recipient and deprive other members that might be interested in your topic. Someday another nube might be searching the forum and find their holy grail in your topic.

Just so you know.. My transition from tubes to transistors wasn't painless. Compounding my difficulty with understanding them was the fact that the first BJTs were Germanium, also known as Point Contact devices that were said to be leaky. Hell, I didn't even know what leaky meant. For all I knew they oozed a slimy substance! Further complicating the learning process was the transistors available to the experimenter were factory rejects. If a factory prime was known to leak what was I to expect from these? As though these hurdles weren't enough to make your head explode early Germaniums were PNPs only. Working with a negative supply voltage was the pits too. Everything seemed upside down!

Anyway, I read every damn book I could get my hands on but made very little progress. Hole Current vs Electron Current plus Carriers made my eyes glaze over. All the data I read had diagrams that looked like a cross sectional view of a stomach wall you see on a wall in your doctors office. My Eureka moment came when I stumbled across a paper back book that was probably no more than 100 pages. It was a long time ago (probably about 1958) but I think it was titled "Repairing Transistor Radios". It wasn't written by an engineer, college professor or any such academic. It was written by an experienced Radio-TV repairman. The preface of the book sold me. Paraphrasing, it read like this....

This book was written primarily for technicians that are finding the transition from vacuum tubes to transistors a rocky road. If you've read about hole current, carriers, N & P junctions, that's fine but we won't be covering any of that here. You can think of hole current as bowling balls if it makes you happy. It's not necessary to know what's going on inside a transistor to understand how they work, what to expect form them and how to successfully troubleshoot and service them. A transistor is a three terminal device. For this reason I will be treating it using tried and true "black box" techniques .................

Happy Trails!

Chris

 

[Solidus]

Well, I wired everything and inserted the circuit, and to put it into terms, it doesn't work. The audio comes through, but the quality is poor. The bass registers aren't transferring through.

I figure I'll design a new one and build everything from scratch. The number of small faults this project has had has run up and down my nerves. I'm not giving up on this build, but I hustled through it to be able to show it to my best friend (and roommate) on my birthday this Wednesday.

Although I have new knowledge and more experience, I have no idea about where to start troubleshooting. I have identified a couple problem areas though.

First one is that the switching supply is generating noise being picked up by the high-impedance and microphonic valve, which is feeding that back into the audio line.

Second is a mistake so incredibly noobish of me I don't even want to post about it. I will post a photo of the circuit up on bread, and the first one of you to correctly guess what it is will get a free dinner from me if I'm ever in your hometown. It is one of the most idiot things a noob could do.

Oh and Chris, if you ever find a reference to that book, that preface speaks kind words to me haha!

 

[Solidus]

 

[CDRIVE]

You need to post much more information so we can diagnose your problems. I hope your not using that protoboard with long jumpers to the tube. Also. many protoboards don't have any insulation on the bottom side. They shouldn't be sitting on a metal object. Multiple nodes could short out.

Tubes are very high Z devices. Hi Z devices can do unexpected things on Protoboards.

Chris

 

[Solidus]

You need to post much more information so we can diagnose your problems.

No need, Chris. I figured it was time to be a big boy and not a nube so I did it myself. I caused the error, I will take responsibility for it and go about debugging or problem attribution.

It is in part to three problems, all of which I will detail below.

They shouldn't be sitting on a metal object. Multiple nodes could short out.

It's not. That's just where I set it for photographing.

Tubes are very high Z devices. Hi Z devices can do unexpected things on Protoboards.

Chris

Problem 1: The metal slots in the breadboard have relative to one another, capacitance, creating a rudimentary, unintended high-passing filter. That drowns out and damps the low-frequency response of the unit.

Problem 2: The HF noise generated by the switching supply is interfering with the valve and generating unintended noise in the signal path.

Problem 3: The long lead wires going from panel, when coupled directly or indirectly to a high-Z amplifier, are pulling radio-level high-frequency crap out of the air and amplifying it. I built myself a bunch of unintended antennae.

Put all three together and you get a mess.

Problem 1 can be corrected near immediately, as I only put the breadboard on circuit because I was pressured for time. I already have a board readied for soldering. 2 can be corrected by building the supply you showed me and decoupling the plate lines. 3 can be lessened by ripping apart the entire panel assembly and rewiring everything, putting everything on shorter leads.

I've spent the past month and a half straining my head over this project so I am going to take a break. There's only so much beating my head against a wall before I lose my wits over this thing. The next valve amp I assemble will probably be started in the next month and it will be done right.

As for this one, left in its current state, I could hope to the amplification gods that the lowest register of the electric guitar is still above the f(c) of the unintentional HP filter that I inadvertently created.

 

[KrisBlueNZ]

Well, I wired everything and inserted the circuit, and to put it into terms, it doesn't work. The audio comes through, but the quality is poor. The bass registers aren't transferring through.

A quick way to check a lot of things is to measure the DC voltage on the anode. The valve should be conducting enough to pull the anode voltage down below the supply voltage (because the anode current causes a voltage drop across the anode load resistor). Typically the anode voltage should be between 50% and 75% of the supply voltage, but anything between about 30% and 80% should work.

I figure I'll design a new one and build everything from scratch. The number of small faults this project has had has run up and down my nerves.

Good idea. It's normal to solder the components straight to the valve socket. I would avoid breadboard because of the stray capacitance.

I'm not giving up on this build, but I hustled through it to be able to show it to my best friend (and roommate) on my birthday this Wednesday.

Happy Birthday in advance

Although I have new knowledge and more experience, I have no idea about where to start troubleshooting. I have identified a couple problem areas though.

First one is that the switching supply is generating noise being picked up by the high-impedance and microphonic valve, which is feeding that back into the audio line.

That's the problem with switching noise. It's wideband, so it's hard to filter out.

For this application it might be best to use a power supply based on a 50 Hz transformer, as you drew up before. This might improve the sound quality too, since the valve will be operating with an anode supply of just 17V instead of just 12V!

Second is a mistake so incredibly noobish of me I don't even want to post about it.

Yeah, don't worry about it! Making mistakes is great, because it forces you to learn how to work out why something isn't working like you expected

 

[KrisBlueNZ]

The metal slots in the breadboard have relative to one another, capacitance, creating a rudimentary, unintended high-passing filter. That drowns out and damps the low-frequency response of the unit.

Actually, capacitance between the signal path and ground attenuates high frequencies, and the problem is especially bad with valves because of their high operating impedances.

For example, 100 pF of stray capacitance at 10 kHz has a reactance of about 160 kilohms, and would cause noticeable attenuation of high frequencies. So it is important to get away from that breadboard.

There must be another problem causing loss of bass frequencies. What are the characteristics of the amplifier you're driving? What type of amp is it? Do you know its input impedance? Loss of bass frequency response with an AC coupled amplifier can be caused by the coupling capacitors not having enough capacitance, and/or by input impedances being lower than they should be. Both have the same effect - the response rolls off as the frequency drops. The low-frequency rolloff is SUPPOSED to occur BELOW the lowest pitch you want to be able to reproduce.

The long lead wires going from panel, when coupled directly or indirectly to a high-Z amplifier, are pulling radio-level high-frequency crap out of the air and amplifying it. I built myself a bunch of unintended antennae.

That sometimes happens with high gain amplifiers. You can add small inductors and capacitors at the input to filter out RF noise, and build a metal cage around sensitive components. Google sensitive ampifier input screening radio interference for guidelines. Also make sure your soldering is fresh and bright, since dry joints can behave like diodes and "detect" (demodulate) strong A.M. radio stations nearby.

 

[Solidus]

Kris, your sense of optimism even when stuff goes wrong makes me feel better and gives me a sense of hope

I haven't added the amplifier module for driving the speakers yet. Aside from the problems at hand, remarks I had were that the valve pre-amp stage is noticeably weak. It has a hard time driving headphones at an appropriate level, even with gain and volume at full-tilt.

Since the second build will be a newer, modified, better reiteration of the first, I'll start posting some of the things I plan on doing to make it (hopefully) better:

I will be ordering blank, copper-clad PCB and etching it myself. The device will be built in an old PC power supply case.

Won't contain speakers, it can always be sent to a speaker setup for that.

I also plan on having three valves paralleled (one for stereo, one for left and right) sent to two transistor amplifier stages for power.

All on one board.
Schematics when I find a way to get my scanner working.

 

[KrisBlueNZ]

Ah, if you're trying to drive headphones directly from your preamp, that will explain why there's no bass. That circuit is a preamplifier; it's not designed to drive speakers or headphones directly. Its output impedance is high, so its output signal has very little power. It has to be used with a power amplifier.

 

[Solidus]

So does that also explain the lack of quality?

Will chaining through to a larger amp unit remedy this?

 

[CDRIVE]

So does that also explain the lack of quality?

We covered this early in the thread. I thought you understood it. This amplifier is designed to drive a PA with an input Z >=100K Ohms. It's not designed to drive low Z earphones. It will drive a crystal earphone though but they don't have good low freq response.

Just for the hell of it measure the resistance of your earphones with your Ohmmeter and post the reading.

Chris

 

[Solidus]

I will admit I don't have a good understanding of impedance in the first place. I'm sorry, I probably should have voiced that earlier but in my nubeness didn't think it was a critical issue.

I do understand that it is the AC analogue of DC resistance, but exactly how it plays into circuits, I have no idea.

My headphones are coming in anywhere from 30-35 ohms. I can look up the specs, I hazard them to be 32. The needle was bouncing around making it hard to get an exact reading.