Thinking about building a valve amp... / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Thinking about building a valve amp...

previous 1 2 3 4 next last

Move Thread

LAN_403

radiotech

Thu Sept 02 2010, 02:56PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Triode -Pentode and triode - sheet beam tetrodes are by far the most common output circuit ever used in non class-AB cirrcuits.
For a dedicated load, ie specified loudspeaker as in a radio or a tape recorder, they are very stable and the effects noted are almost rooted in failure of electrolytic capacitors, particularly the multi section cans or tubular. The other issue is the final tube deterioration with age and heat and gas. The cathode bypass, is shaped to limit the lows (for a small speaker).
The tubes in this project are the wild card. The condition is unknown and wont be found out until tried. As mentioned before, it is essential no grid current flows due to leaky coupling. One of the first things a tech would do in servicing an amplifier with complaint of distortion afeter a period is to hang a VTVM on the grids of the output and watch for creep-up. The triode plate resistor could increase to 4 times its original value and still work (poorly
)
From the advent of stereo (1957 to the acceptance of solid state in home entertainment (1968) most low end stereos used these circuits and a good living was made keeping them running.

As to the distortion, this amplifier project is simply going to give the sound of that era. Recall the fiasco that ensued when the first transistor amplifiers came out with amazingly low distortion but people just couldnt accept the sound, especially the high damping.
It became necessary to invent TIM , transient intermodulation distortion, to seperate the various sounds of the amps.

As to howling and siren sounds, thats why there were shock mounted tube sockets for such devils as ECC83 and EF86 types. In the old Grundig TK 46 taperecorder, you could unplug the microphone and yell into plug socket and hear it on the tape.

One great resource is the Wireless World magazines from that era.
I remember a monthly column called 'Grid Leak ' by a writer called Unbiased.

Mads Barnkob

Thu Sept 02 2010, 04:27PM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

For inspiration you could look at my single end tube amplifier using a UCL82 tube, which is very similar to your ECL83 tube :)

Proud Mary

Thu Sept 02 2010, 05:07PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

radiotech wrote ...

Triode -Pentode and triode - sheet beam tetrodes are by far the most common output circuit ever used in non class-AB cirrcuits.
For a dedicated load, ie specified loudspeaker as in a radio or a tape recorder, they are very stable and the effects noted are almost rooted in failure of electrolytic capacitors, particularly the multi section cans or tubular. The other issue is the final tube deterioration with age and heat and gas. The cathode bypass, is shaped to limit the low

You are speaking from the position of someone servicing a piece of kit designed by someone else, and hopefully not marketed until stability and other problems have been ironed out.

It seems to me that the situation of someone like Avalanche who has never once before designed a valve circuit is rather differrent to that of the repair man, and that he would do well to study the sources of instability in amplifier design, if he is not to produce the kind of farmyard chorus of unwanted special effects I have described.

Avalanche

Fri Sept 03 2010, 05:16PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

Thanks for the input - I'm taking everything into account but I've been busy working on the case today so that everything stays on one piece

Successfully re-wound the transformer last night, this time with 32SWG wire and an insulating layer of brown paper every 100 turns. 1534 turns (again) and this time no break down - it seems solid. Anyway more on that later.

A sneak preview of progress so far... I built the case from scratch a few years ago for a theremin, but I've modified it for the valve amp. The wood got a bit messy and inconsistent after I modified it, so I sprayed it matt black.

radiotech

Fri Sept 03 2010, 06:31PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Thats good to see the core orientation of the power transformer perpindicular to the output transformers for minimun hum coupling. Another thing from TV days was to add a shading band since crts were very sensitive beam hum.

Steve Conner

Sat Sept 04 2010, 11:27AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Nice looking amp!

I don't know what Proud Mary is talking about. Those motorboating and squealing effects are caused by feedback, not cured by it. Tube amp circuits are so high impedance that you have to be very careful with layout. Even a few pF of capacitance between wires can make it squeal (but never motorboat- motorboating is caused by time constants far too long to be associated with stray capacitance)

You'll want negative feedback anyway, modern speakers sound terrible when driven by a pentode amp without any.

Try searching for the Mullard book of circuits, there should be an example one for the ECL8x tubes.

I've been running this

in my hi-fi for the last year. I'm now using GEC KT88s, I got six of them from two scrapped Brandenburg HV power supplies, and managed to make two matched pairs.

Proud Mary

Sat Sept 04 2010, 01:02PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Steve McConner wrote ...

I don't know what Proud Mary is talking about. Those motorboating and squealing effects are caused by feedback, not cured by it. Tube amp circuits are so high impedance that you have to be very careful with layout. Even a few pF of capacitance between wires can make it squeal (but never motorboat- motorboating is caused by time constants far too long to be associated with stray capacitance)

You'll want negative feedback anyway, modern speakers sound terrible when driven by a pentode amp without any.

I wrote above:

"I see that Mullard give Dtot 10.5% for ECL83, and though you say you don't mind a bit of distortion, it's as well to remember that triode-output pentodes in radios and televisions very commonly used a simple negative feedback loop from the output transformer secondary to the triode cathode line both to reduce distortion and unwanted input/output coupling consequences - "motorboating," gain peaking, hound-like baying, and air-raid siren effects - to which the casually configured triode pentode is prone."

Avalanche

Wed Sept 08 2010, 07:27PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

Well I have been making some progress over the last few days, case and layout is complete and everything is fixed down so I can work on it properly.

First slight mistake is that I totally underestimated how poor the regulation would be on the transformer, so I'm a bit down on heater voltage, and HT - but that's ok because I can fairly easily add a few turns for the heaters (they are only running at about 5.8V so emission will be right down I'm guessing - I don't know how critical it is). HT hopefully won't be such a problem - my DC rail voltage is at 190V under load, as opposed to 200V.

I went for a choke input filter after the rectifying tube, because the (oldish) book I am reading at the moment seems to think choke input is the best. Finding the right inductor was a challenge, as it has to have such a high inductance - eventually I found a small mains transformer that measured over 8H on the primary. I had to somehow prevent it from saturating, so I took a hacksaw and cut 2 slots in each side, which reduced the measured inductance to about 5H. I used a 100uF electrolytic after this, and the result is that the ripple is down to about 1V peak to peak under load. Not bad I thought?!

Anyway, I looked up the correct self bias cathode resistors, fitted them to one of the triodes and one of the pentodes, and connected them both as class A amplifiers. Connected an MP3 player to the input, loudspeaker to the output transformer and powered on. Slowly, up came the HT and I could hear something in the speaker. It wasn't music. As the HT came up, the noise got louder and louder. I could only describe it as a warehouse full of angry cats, screaming children, and squealing fan belts. As it had now reached a volume that I wouldn't expect a single valve to be capable of, I turned off the power...

Fitting a 1nF capacitor between the pentode anode and the triode cathode seems to stop it, but it severely limits the frequency response of the triode stage (peaks at about 120Hz, and cuts off at about 500Hz!). The pentode stage alone peaks at 8Khz and seems to have a reasonable response. Anyway, I've been playing music on the amp and it is a reasonable volume for 1 channel of class A.

So that's where I'm at right now. I'm going to get back to the drawing board and try and design this thing properly as a class A amp, starting with the triode preamp stage and then coupling it properly to the pentode with some proper negative feedback somewhere. As it's unlikely I'll ever turn it into a class AB, I'm going to probably use the remaining 2 valves as further preamps for the guitar (I think the guitar will need at least 1 more triode stage to have any chance of getting any overdrive).

This should probably have been a project thread afterall ... oh well!

radiotech

Wed Sept 08 2010, 09:47PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Here's a layout suggestion for units with choke input filters.

Avalanche

Thu Sept 09 2010, 07:19PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

Thanks for posting that, I've just realised I need to rotate the core of the choke by 90* as I just screwed it inside the case without thinking. I'm guessing the small amount of mains hum I'm getting is being caused by a combination of that plus my heater center tap no longer being in the center (I've had to add turns, but I can sort that out). The hum is not reduced at all by any amount of negative feedback, so it must be coupled in from the power supply or the heaters.

I've just been playing around with negative feedback, what would be the pros and cons of taking the feedback from before or after the AF transformer? I've tried both, just simple resistive only feedback with a decade box, and I thought feedback direct from the pentode seemed cleaner. I would think that taking it after the AF transformer would bring in problems with phase shift, and make the frequency response a bit less linear... I can see myself getting a bit hooked on feedback networks made with various components to try and get some nice tones.

I now need to come up with a quick way to do frequency response sweeps and get them into a spreadsheet!

previous 1 2 3 4 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob