First valve amplifier build

[Gabe001]

I was wondering if I could get some views on a little side project of mine. I'm attempting to build my first valve amplifier which is proving a bit of a steep learning curve

Unwilling to spend £££ on one and hoping to enjoy the building as much as the listening, I found a cheap chassis online from a radiogram (I think) which I thought could form a sound basis for a homebrew job. Grim Joeseph and Jerry (cathoderay57) have already helped me out quite a bit, but it would be very interesting and informative to get opinions from here

The original amp circuit which I traced out is below. I've got it going with an HT of around 235v after the rectifier and it's got an unacceptable degree of hiss. AC hum is negligible. The smoothing cap is 100uf, 400uf and 16uf and seems healthy.

The proposed modifications are also attached. The valves are 6060, 6c4 and 2x6k6gt (original). I've removed the bass and treble controls in my modified version and set up the 6k6s in P-P

I am not an audiophile by any means, so I'm not looking for acoustic nirvana or to annoy the neighborhood, just enjoyable sound reproduction and reasonable but not uncomfortable volume without distortion. 5 watts nice and clean would be plenty.

The circuit employs the 6c4 as a phase splitter (I know the 6060 would have been better, but the 6c4 is easier to rewire and the valve is close to the 6k6s). I got the idea from a circuit for an amp called the bogen cathodynamite -or rather Jerry found it. I'll also deal with the power supply, earthing etc but I can sort that out myself.

I would be grateful for some feedback. Also, how do you work out the size of the resistors before G1 and the capacitors straddling the cathode resistors (values vary on online circuits between 22 and 1000uf)

Thanks
Gabriel

[PJL]

The 'original' circuit has the output stage as a long-tail pair which is not something I have seen before. Long tail pairs are typically used for the phase splitter. What was the original amplifier, was it amateur construction? The components are a real mix of vintage parts that most likely will be faulty.

The 'new' design has an error as the Bogen circuit uses a concertina phase-splitter but you are missing the cathode resistor.

For a first build, I would recommend you go for a trusted design, the Mullard 5-10 perhaps, and you need to observe layout rules or it will be unstable.

[Radio Wrangler]

Your circuit also has a half-wave HT rectifier. This creates a DC current component in the mains transformer which can cause trouble if the transformer wasn't designed for it. It also means that your reservoir capacitors only get topped up half as often as with a full wave rectifier. The three extra silicon diodes to make a bridge rectifier are cheap and will make life easier for your power supply components. Less hum, too.

For a first amplifier, you should definitely stick to a reputable, existing design. The Mullard ones are amongst the best. Even with these there are pitfalls for the inexperienced.

David

[Edward Huggins]

An unusual choice of valves, half-wave rectification and no negative feedback. An unusual circuit that puzzles even those who know. You've kind of created a problem for yourself, but at the same time I don't want to deter you. This will need re-working and I can't predict the sound quality from this as it is. Cathode caps are usually around 25uf. Why not change the output pair to 6V6GTs and base your design on these? There are various 10W amps based on the good old 6V6 to guide you.

[Sideband]

Originally Posted by PJL

For a first build, I would recommend you go for a trusted design, the Mullard 5-10 perhaps, and you need to observe layout rules or it will be unstable.

Originally Posted by Radio Wrangler

For a first amplifier, you should definitely stick to a reputable, existing design. The Mullard ones are amongst the best. Even with these there are pitfalls for the inexperienced.

David

Couldn't agree more. My first 'proper' amplifier was a 5-10, 5 valve 10 watt...in fact two of them for stereo. Worked first time built to the layout suggested by Mullard and also using the correct output transformers. Served me well for many years and I passed it on to a friend who also used it for many years. I subsequently built several more for friends.

A good 'starter' amplifier from the Mullard stable would be a '3-3'. It's a three valve three watt amplifier which might sound a bit weedy for power but three proper watts into an efficient speaker is LOUD! Quality is excellent and I have built several of these in the past including one fitted to a record player. Two make an excellent stereo amplifier. You need to be careful with the layout and wiring but if you follow the Mullard diagrams it will work first time without becoming an oscillator.

In any high quality amplifier it's essential to use good quality (but not audiophile) parts especially around the pre-amplifier stages otherwise it can result in unwanted hiss and noise.

[Robert Gribnau]

The anode of the first triode section of the 6060 is connected straight to B+, so the anode resistor is missing. Without is no signal will pass to the second stage.

Maybe intentional but the volume pot of input 2 will also be a second volume pot for input 1. If the plug for input 2 shorts the hot pin to ground when no plug is inserted, that volume pot will work 'strange' (first the volume goes up, but from some point going the volume goes down and becomes zero again at the 'maximum' setting).

The anode resistor of the second section of the 6080 and the screen grids of the 6V6GT's are not connected to B+.

And (ofcourse) the cathode resistors and bypass capacitors of the 6080 are not there yet.

[Gabe001]

Hi Paul, thanks for the feedback.

I put the cathode resitors in, it was an oversight (past my bedtime). The draft is attached.

Link to bogen build here: http://ppamps.blogspot.com/2013/03/?m=1

I could use the 6060 (=ecc81) as a phase splitter without much difficulty and use the 6c4 as an input triode or even substitute the valves (ecc83/ef91) to get more gain, which would put it (broadly) into Mullard territory, but the chassis I've got doesn't lend itself to the Mullard 5-10 layout with regards to the position of the output transformer and valves, so I am concerned about stability and hum or hiss if I push it too much. It'll be used with a headphones level signal (Bluetooth receiver) not a magnetic cartridge, so I need to tone it down a few notches with regards to gain anyway.

I know nothing about its provenance. I was told radiogram which would explain the 2 inputs.

Can I double check that a single common cathode resistor in a push pull circuit of 400 ohms (recommended for 6k6) equates to 2 separate ones of 800 ohms (820 is the closest match)? I have no way of knowing if the output valves are matched

Thanks
Gabriel

[Diabolical Artificer]

Can I double check that a single common cathode resistor in a push pull circuit of 400 ohms (recommended for 6k6) equates to 2 separate ones of 800 ohms..

That's right. To find Ck (cathode bypass C) use the formula 1/(6.28 x R x F) F = frequency. What frequency you may ask? There is a complicated answer and a simple one, which choose a low frequency that sounds about right, say 15hz, so 1/(6.28 x 400 x 15) = 2.6 to the minus 5 or 0.000026 farad's or 25u.

The nearest common value is 22u. Having wrote all that it's easier to tack in whatever cap you have lying around and give the amp a listen, then change if not right. More bass, use higher cap value. Lastly make sure your cap has sufficient V rating. It has to be higher than Vk, 3 times is about right, check here for more details - http://www.valvewizard.co.uk/se.html

Lastly if possible I'd use the 6060 as a LTP phase splitter and the 6C4 as input, though it depends on how much gain you need.

Andy.

[Herald1360]

Some grid leaks on the output pair would help define their anode currents! Use around 330k. Stay with the separate cathode bias arrangement, it's much safer for the valves especially if they're a bit tired and mismatched!

Choose the anode and cathode tail resistors of the 6C4 to drop about a quarter of the HT each at the anode current it runs at. Take its grid leak off ground and return it to the junction of its cathode bias resistor and the long tail one otherwise the valve will just be cut off! 1M will do for the value.

Setting it up (the phase splitter) is a bit tricky- first, what's its HT supply? Next use the datasheet to select the bias resistor to give about 10mA anode current at half the HT voltage then choose the anode and tail resistors to drop 25% of HT at 10mA. Then try it and tweak the bias resistor as required!

Have fun, you'll learn a lot!

[Sideband]

Worth noting that the 6C4 is basically half an ECC82.

[cathoderay57]

Most of the 6C4 concertina/cathodyne phase splitter circuits including the Bogen use a high anode voltage supply circa 300v presumably to get sufficient voltage swing to drive the output valves. Conversely the 6K6GT valves want typically 285v on their anodes and grid 2s. Some juggling about with the HT supply is needed if this is to work efficiently. To begin with, how much AC voltage have we got from the mains transformer off-load? All things are possible including use of a voltage doubler for the HT supply. Cheers, Jerry

[Gabe001]

so many comments appeared at once, I could only see one when I replied to PJL this morning

Thank you for the feedback, much appreciated. I have corrected the glaring errors with the diagram i.e. cathode resistors, grid leak resistors and so forth. I'll post a diagram update soon enough. I don't have much experience with negative feedback from speaker, the only time I came across it was in a Philips 462a. If its the same concept, by and large, I'm happy to replicate it using the 462a as a template.

I didn't realise that volume pot 2 affects input 1, but its obvious now that its been pointed out so I think a single input is the way forward. I don't think i need 2 anyway. Thanks for letting me know, it had completely escaped me.

HT is still unknown. Currently, with the old rectifier, I get 230v DC at the 400uf smoothing cap, which is rated 350v working. A bridge + suitable resistor is an option, but its been pointed out to me that a 50Hz hum may be less intrusive than a 100Hz hum to the ear and that a bridge may not lead to a significant improvement other than raising HT. In any case, I've got all I need to do some experimentation. Using a bridge rectifier means that I can use modern caps with more "obtainable" values than 400uf, should this fail in future, and higher voltage ratings.

Regarding emulating a mullard 5-10 or 3-3, the idea of using the 6060 as a phase splitter and the 6c4 in mullard 5-10 style (potentially substituting the 6060 to ecc83 and the 6c4 to ef91) is appealing to me, as I do already have the valves available. My concern would be the layout of the chassis as I don't have the tools or confidence to make substantial modifications in the metal sheet itself e.g drilling new holes for valves and the OPT. I understand the Mullard layout should be copied as faithfully as possible. As such, attempting the Bogen style amp first with the anode and cathode biased 6c4 is reasonably straightforward-ish and completely reversible as an initial attempt and more in keeping with the original which I presume worked satisfactorily at some point. So the plan, so far, is to try the Bogen style amp first, see whether this works, and then the Mullard style amp second.

So why didn't I use 6v6's (yet)? 6v6s draw a bit more current than the 6k6s. Marginal I know, but the output transformer is an unknown entity and I'm not sure what current it can and cannot handle.

Any yes, I do know that I made things hard for myself. It wouldn't be half as enjoyable it it were easy, and I want to learn from the process, so happy to do some sums, experiment, and start again from scratch if I have to.

Gabriel

[Gabe001]

Originally Posted by cathoderay57

Most of the 6C4 concertina/cathodyne phase splitter circuits including the Bogen use a high anode voltage supply circa 300v presumably to get sufficient voltage swing to drive the output valves. Conversely the 6K6GT valves want typically 285v on their anodes and grid 2s. Some juggling about with the HT supply is needed if this is to work efficiently. To begin with, how much AC voltage have we got from the mains transformer off-load? All things are possible including use of a voltage doubler for the HT supply. Cheers, Jerry

241 AC RMS off load from transformer. OPT DC resistance 400ohms per section (CT to anode)

[GrimJosef]

Originally Posted by cathoderay57

... Some juggling about with the HT supply is needed if this is to work efficiently. To begin with, how much AC voltage have we got from the mains transformer off-load? All things are possible including use of a voltage doubler for the HT supply ...

Looking at the photo and the fairly primitive original design I'm afraid that the manufacturer may have spent little on the mains transformer, and it may be running quite close to its current limit. The 6K6's do seem to need significant current, albeit at relatively low voltage, to work as they were designed. So while it might have been nice to boost the HT voltage with a more up-market rectifier arrangement I was nervous that any consequential impact on the safely available current could be serious. I think I'd want to know what the mains transformer's current limit was before setting about redesigning the audio amp's output stage (other than to do away with the LTP scheme, which does need to go).

Cheers,

GJ

[Diabolical Artificer]

Don't worry about layout too much, as long as your grounds are sorted you should be fine.

Andy.

[cathoderay57]

Hi Gabriel, I note your comments about wishing to avoid cutting and drilling. However, if you were prepared to do a little bit of "chassis-bashing" you could have the attached mains transformer FOC which has more heft than the one you've got fitted and could be installed similarly as a drop-through. It would give you a lot more headroom for the HT and has enough oomph in the heater windings having come out of an instrument containing 20 B9A valves. It's fairly heavy and while I could post it to you at cost I could also bring it to RetroTech. Would need to be mounted so that the laminations are 90 degrees to those of the OPT. Cheers, Jerry

[Edward Huggins]

I'm not sure that the original chassis now being worked on, was ever from a Radiogram?

[Gabe001]

Righto, I've made some progress.
The half wave rectifier was concealing the rectifier valve socket hole. A closer look at the transformer revealed a cut wire. It turns out that the transformer's HT secondary winding was originally centre tapped to work with a rectifier valve.

So I lashed up 2 diodes on a tag strip with a 100ohm surge limiting resistor. The transformer appears comfortable but I'm keeping a close eye on it. The lamp limiter behaved as it should

HT is as follows:
279v before 100ohm (5w) surge limiting resistor
271 after the resistor equating to 80mA total current draw
263 at anode of 6k6 (25mA current draw)
230v at the screen of the 6k6

That's it for today. I'll try to experiment with a temporary arrangement of the output phase tomorrow with either the 6060 (ecc81) as a phase splitter or the 6c4 depending on the advice received.

Thanks again

Gabriel

[Gabe001]

Originally Posted by cathoderay57

Hi Gabriel, I note your comments about wishing to avoid cutting and drilling. However, if you were prepared to do a little bit of "chassis-bashing" you could have the attached mains transformer FOC which has more heft than the one you've got fitted and could be installed similarly as a drop-through. It would give you a lot more headroom for the HT and has enough oomph in the heater windings having come out of an instrument containing 20 B9A valves. It's fairly heavy and while I could post it to you at cost I could also bring it to RetroTech. Would need to be mounted so that the laminations are 90 degrees to those of the OPT. Cheers, Jerry

That's very kind. PM'd you

[Robert Gribnau]

Originally Posted by Robert Gribnau

Maybe intentional but the volume pot of input 2 will also be a second volume pot for input 1. If the plug for input 2 shorts the hot pin to ground when no plug is inserted, that volume pot will work 'strange' (first the volume goes up, but from some point going the volume goes down and becomes zero again at the 'maximum' setting).

What I wrote is probably wrong. If the values of the resistor coming from the wiper of the second pot and the resistor coming from the first stage are large enough (like 470K) than the pots will not influence eachother much.