50L6 guitar amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey guys,

I'm new to the forum, but I've learned alot from reading on here.

I've done some electronic projects including many guitar pedals, I know schematics and circuits well enough, I'm no stranger to DIY.

I've built 2 tube amps from scratch so I have moderate amount of tube experience, quite good soldering skills and I know what not to do.

I'm looking to built a new AA5 based guitar amp and want to use as much parts as I can from my junk box.

Here's what I found in there :

- An isolation transformer 60VA
- Power tubes: 50L6
- Preamp tubes: 12au7, 12ax7, 6n2p-ev, 6av6, 12av6.
- A nice metal box which can be use for the chassis
- All the stuff to built an eyelet board
- Some usable sockets

2 parts are missing though, the output transformer.. and I don't have any 35Z5.

My first idea was to built a 50L6 PP amp, one 12ax7 at the preamp, but instead using a SS rectifier at the power supply.

But I think the 60VA transformer has not enough juice to power all the filaments... then I found this schematic using only one 50L6 (from Tino Zottola's book) :

http://www.kirtland.com/guitar1/guitar1/MA1_files/50L6_Amp-1.gif

Seems like a nice little amp to built, my isolation transformer has multiple primary wires (COM-120-208-230-460), I'm pretty sure by using the COM/230 wires it will fall near the perfect 60VAC at the secondary.

Okay now is this the correct way to 'guestimate' the things?

50L6 filaments -> 150 ma
Max plate -> 50 ma
Max Screen -> 10 ma
12ax7 filaments (12V serie) -> 150 ma
Max plate -> 2.4 ma

Roughly 362.4 ma, so the 60VA isolation PT should be okay?

Am I correct saying this design uses an half wave rectifier circuit (kinda doubler).. I wonder if this could cause some heat in the power transformer, would it be better instead to use a full wave bridge? I read somewhere that power transformers are designed to be symmetrically loaded.

On the other hand, the plate voltage suggested in the schematic seems a little low?

My technical knowledges are not near of what I hope they should, trying to figure out my options.

I would really appreciate some advices and comments from the more experienced people.

Thank you!
 
It's little money for a Triad N-68X, which should be perfect for the project. That "iron" gives you 400 or so mA. RMS. Subtract the 150 mA. the series heater string uses and 250 mA. RMS are left. So, full wave bridge rectifying the B+, with 4X UF4007s, will easily provide 125 mA. for the signal tubes. That's plenty for both SE and PP concepts.
 
A pair of 6L6s can give you 50 Watts of audio but a single 50L6 in class-A is about 4 Watts audio. 50L6 was made in order to avoid a power transformer in small radios and the price was super bad efficiency. You can not make a power-transformer-less amp that is safe to connect to anything else, not a problem for an AM radio. I did that many, many years ago and burnt a chair when it touched the very hot 50L6. The results are not worth anything so it's only about how you want to waste your time.
The most important thing about guitar amps is a very high impedance input. There is no real advantage to tubes if you understand what to avoid.
 
@Eli

So the 60VA iso transformer I already have on hand should be more than enough. Here it is.

That's what I thought about using uf4007 fast recovery diodes bridge, I used them in my 18W PP el84 built. I'm working on a 50L6 schematic, will post it later. Thanks for the reply Eli

@steveu Thanks for the advises, yep I use to have some 50/100W in the basement, which are collecting dust btw.. I love cranked little amps for home practice (I don't like attenuators). I think with an iso transformer they are safe to play, right? I already owned some 35w4/50c5/12ax7 old canadian amps and love them, they all have been modified with an iso transformer, keeping the tube rectifiers just like the suggested modification on the instructable website. I'm not looking for performance, just having fun in the diy world, trying to learn and recycle things as much as possible.

The most important thing about guitar amps is a very high impedance input. There is no real advantage to tubes if you understand what to avoid.
Sorry I don't understand, could you please explain what it means.

Thanks again guys!
 

PRR

Member
Joined 2003
Paid Member
...a 50L6 PP amp, one 12ax7 at the preamp...

"PP" means two 50L6 in push-pull?

You need 2 gain-stages and a phase inverter--- highly suggests *two* 12AX7.

You have two 50V and two 12V heaters: 125V, can be fed right from a 120VAC winding.

Taking 100mA for 2*50L6 (ignore the little bottles) needs around 170mA for the DC supply. Add 150mA for the heater string, makes 320mA @ 120VAC or 39VA. A 60VA 120V:120V iso-former seems suitable. (Your many-tap transformer really confuses the issue and _I_ would just use a simple 120:120.)
 
Many thanks for the reply PRR

Sorry my bad, I understand better now. I was figuring 300ma for the filament chain which is wrong, they are hooked in serie! Indeed the PS is suitable for a 50L6 pair.

What you say confirm to me that I can built this little amp 'as is' without having to modify the power supply section, the 60VA should be good enough to work without overheating because of the half wave rectifier. I mentionned about the multiple primary tap because by using COM/230 at the primary, I get roughly 68 VAC at the secondary, which is perfect for the 50L6 SE project.

I have 2 other iso transformers which are exactly the same, I would also love a paralled 50C5 or 50L6 SE but didn't find any schematic. For now I have some reading and learning to do I think, maybe the next built!

I'm gonna make this one just like it is in the schematic and then after see what I can mod to my taste.

In a couple of lines you save me weeks, thanks again Sir! :)
 
(Your many-tap transformer really confuses the issue and _I_ would just use a simple 120:120.)

Sorry, misunderstood again..

Now I see that by using a multi primary tap iso transformer I'm gonna deal with heat problems because of the larger wire cross section, versus a simple 120:120.

I did an intensive google search in regard with current loss and multi tap transformers but didn't find anything that I can understand.

By using my multi tap transformers, even on the 1:1 winding, I mean COM/120, it will still be worst (current loss) than a simple 120:120, right?

I'm gonna do some heat test.

Thanks again all, for pointing me in the right direction! :)
 
Be aware of possible hum when feeding the heaters from the same secondary winding as the HT supply and minimize it: Provide a Greinacher voltage doubler as HT source which allows you to ground the secondary. Then build the heater chain as follows: 12AX7-12AX7-25L6-25L6, with the input triode's heater pin #4 or #5 (you'll need a close look) connected to the grounded end.
Best regards!
 
...I use to have some 50/100W (guitar amps) in the basement, which are collecting dust btw.. I love cranked little amps for home practice
I agree with you there. I have two 15 W valve amps, and those "little" amps are far too loud to use at home. I have a DIY two-watt valve amp, and even that one is too loud to use at home unless I keep it well turned down, or use a DIY attenuator.

At the moment, I'm experimenting with a guitar amp design using a single 6JW8 triode-pentode. The valve cost me one buck (USD), and the amp puts out about 1/4 watt of audio power - which seems to be plenty loud enough for use in an apartment.

I've never used a 50L6, but I've looked at the datasheets. It is extremely non-linear, which is one reason why the valve Hi-Fi crowd despises it. But for electric guitar, "very non-linear" might be exactly what we want to get a nice rich clean tone. I look forward to hearing more about your project.


-Gnobuddy
 
Be aware of possible hum when feeding the heaters from the same secondary winding as the HT supply and minimize it: Provide a Greinacher voltage doubler as HT source which allows you to ground the secondary. Then build the heater chain as follows: 12AX7-12AX7-25L6-25L6, with the input triode's heater pin #4 or #5 (you'll need a close look) connected to the grounded end.
Best regards!

Thank you Kay, didn't know about the Greinacher voltage doubler.

Okay I see, always thought that there was no polarity to filaments.

I agree with you there. I have two 15 W valve amps, and those "little" amps are far too loud to use at home. I have a DIY two-watt valve amp, and even that one is too loud to use at home unless I keep it well turned down, or use a DIY attenuator.

At the moment, I'm experimenting with a guitar amp design using a single 6JW8 triode-pentode. The valve cost me one buck (USD), and the amp puts out about 1/4 watt of audio power - which seems to be plenty loud enough for use in an apartment.

I've never used a 50L6, but I've looked at the datasheets. It is extremely non-linear, which is one reason why the valve Hi-Fi crowd despises it. But for electric guitar, "very non-linear" might be exactly what we want to get a nice rich clean tone. I look forward to hearing more about your project.


-Gnobuddy

The 6JW8 looks like a nice tube, similar to 6AN8 often used in hi-fi as output stage driver.

I just finished a 1W firefly amp and I agree that it's far enough power at home.

Yep I have 3 old canadian amps (50C5/12ax7) and they all have a killer sound, I can't wait to hear the 50L6, will post sound track as it is completed :)
 
...didn't know about the Greinacher voltage doubler.
The schematic you posted earlier (for the SE 5L6 amp) actually uses a voltage doubler to generate the B+ for the amp. That's how it manages to turn 57 volts AC RMS into a B+ of 125 volts DC.

The doubler is made up of the two 220 uF caps, along with the two 1N4005 silicon diodes.
Okay I see, always thought that there was no polarity to filaments.
You are right, there isn't!

The problem is a different one. You are feeding nearly 60 volts AC across the two ends of the filament. If you ground one end of that transformer winding, the other end is flapping about wildly, 60 times a second, from roughly (-84V) to (+84V) and back, over and over again.

So when you connect the heaters in series and wire them across this same 60V transformer winding, one end of one heater - the one connected to ground - is sitting politely at zero volts. As you go along the heater chain, the AC voltage keeps increasing and increasing. When you get to the opposite end, you're back to those violent swings from (-84V) to (+84V).

Now, if that very noisy end of the heater chain was powering the input triode (first half of the 12AX7), you would probably have terrible hum in your amp.

On the other hand, if the grounded end of the heater chain is powering that input triode, things will be much better. The noisy end of the heater chain will now try to stir up trouble in the 50L6, but there is a much bigger guitar signal there, and much less gain between 50L6 and loudspeaker. So it won't cause as much of a problem.

The schematic you posted shows the heater wiring correctly - note that the 12AX7 is connected to the grounded end of the 60V AC winding.

Kay Pirinha's point was a refinement of that idea. Inside the 12AX7, there are two triodes. Your schematic uses pin 7 as the input - the most sensitive stage to hum. Pin 8 is the cathode closest to that. We would like the end of the heater nearest to that cathode - the one connected to pin 8 - to be grounded.

So which end of the heater - pin 5 or pin 4 - is physically closest to that cathode? I don't know. You may be able to tell with a magnifying lens. Or you can hope the designer of that amp thought of this already, and that's why he/she shows pin 5 connected to ground. Or you can always try swapping the heater wires going to pins 4 and 5, to see if the amp has less hum one way than the other.


-Gnobuddy
 
Actually there ain't no polarity. But we should try to keep the heater-cathode AC voltage as low as possible for the most sensitive stage.
Best regards!


Edit: Should have been reading Gnobuddy's posting first, as he said just the same before :rolleyes:!
 
Last edited:
Hey guys!

Well, I understand everything until the magnifying lens, you mean the 50L6 cathode? The glass is black coated! If I make sure to place the preamp tube(s) far away from the 50L6 should I worry about this refinement. Is this why I'm hearing hum from 50C5 tubes with H-K leak, the AC coming from the heaters is leaking to the cathode, right?

Ive just finished to tweak my little firefly amp, which had some kind of an oscillation problem at high gain and finally it was an error in the design, I positionned the preamp tubes too close of the OT so the signal was escaping, I dunno maybe from the primary wires or the OT itself, back to the preamp tubes. By using these preamp shield caps the problem is now gone and I'm very happy.

Electronic is full of strange things which make my day every day! But so far I'm still learning :)

Thanks again!
 
No, you should have a very close look at the internal structure of the 12AX7 that you intend to be the 1st tube. There are two triodes (surprise, ain't it?) with two cathodes and two heater filaments. Connect the open heater end (i.e. pin 4 or 5) of the 1st stage triode to ground.


Best regards!


Btw, it's much better to opt for a 12AX7A or ECC83 or 7025 for this purpose!
 
As Kay says, just make sure that one end of the 12AX7 heater is connected to ground. (This means neither end of the 50L6 heater will be connected to ground.)

There is some chance that the amp might hum less with pin 4 grounded. Or it might be pin 5 grounded. If you feel like it, after wiring and testing, you can try swapping pin 4 and 5 of the 12AX7 to see if one is quieter than the other.
If I make sure to place the preamp tube(s) far away from the 50L6 should I worry about this refinement.
It's always good to leave a little distance between the preamp tubes and output tube(s) - but this is mainly to keep the thing from oscillating at high frequencies (it might squeal, or turn into an RF transmitter.)

the AC coming from the heaters is leaking to the cathode, right?
Exactly. The leak can happen because of poor heater-cathode insulation, or simply because of stray capacitance between the heater wires and the tiny metal cathode tube they are stuffed into.

Honestly, in my opinion, it is madness to power the heaters with AC. But decades ago, there was no choice - there was no practical way to get enough clean DC power for the heaters, other than using an actual battery for the job!

So the designers and builders of that time evolved all these careful tricks about heater wiring layout, twisted wires, et cetera, to try and manage the headache they were stuck with.

Things are different now - we can get inexpensive switching power supplies that put out amps of clean DC. I take this route with most of my home-brew valve stuff. But it's not a good solution for this 50L6 project, because you need either separate 12.6V and 50V DC power supplies, or a 63 V power supply, which is hard to get.

(48V DC is relatively cheap and easy to find, for example: https://www.digikey.ca/product-detail/en/mean-well-usa-inc/IRM-30-48/1866-3049-ND/7704671 )

Electronic is full of strange things which make my day every day!
This keeps it interesting, right? :)

My other main hobby is music, and it's the same story there. There is always some new thing to learn about music, or about playing your instrument. Nobody ever gets all the way to the end of the road. There isn't one! :)


-Gnobuddy
 
No, you should have a very close look at the internal structure of the 12AX7 that you intend to be the 1st tube.

Okay! I never tried to identify parts inside a tube, found this and will have a look. Thanks alot for sharing tips coming from long time experience! I feel like this built will be a great success.

Btw, it's much better to opt for a 12AX7A or ECC83 or 7025 for this purpose!

Years ago Ive built an amp (using parts coming from an old radio) based on the Garnet 5p schematic, the hum is 'acceptable' but I could do better, where's the best spot in the chassis to ground the 12AX7?

When I power up this amp, the 35W4 is glowing very bright for a couple of seconds, then while the 50C5A warms up it becomes normal. I found that the A suffix in any tubes designates a controlled filament warm up time, I also found that it's supposed to be a lower noise version, is this true.

I tried to swap the 50C5A for a 50C5 and it's the same, I wonder what could be causing that. The 12AX7 is ECC83S from JJ and the rectifier is labeled 35W4/HY90 made in japan.

As Kay says, just make sure that one end of the 12AX7 heater is connected to ground. (This means neither end of the 50L6 heater will be connected to ground.)

There is some chance that the amp might hum less with pin 4 grounded. Or it might be pin 5 grounded. If you feel like it, after wiring and testing, you can try swapping pin 4 and 5 of the 12AX7 to see if one is quieter than the other.

Yep I get this part, I wonder if it worths to make a ground switch, or maybe the extra wires would be causing more hum, I mean like that

Honestly, in my opinion, it is madness to power the heaters with AC. But decades ago, there was no choice - there was no practical way to get enough clean DC power for the heaters, other than using an actual battery for the job!

So the designers and builders of that time evolved all these careful tricks about heater wiring layout, twisted wires, et cetera, to try and manage the headache they were stuck with.

Things are different now - we can get inexpensive switching power supplies that put out amps of clean DC. I take this route with most of my home-brew valve stuff. But it's not a good solution for this 50L6 project, because you need either separate 12.6V and 50V DC power supplies, or a 63 V power supply, which is hard to get.

(48V DC is relatively cheap and easy to find, for example: https://www.digikey.ca/product-detail/en/mean-well-usa-inc/IRM-30-48/1866-3049-ND/7704671 )


This keeps it interesting, right?

My other main hobby is music, and it's the same story there. There is always some new thing to learn about music, or about playing your instrument. Nobody ever gets all the way to the end of the road. There isn't one!


-Gnobuddy

The only DC heaters I did was on my valve junior V1, with some other mods I can't remember and I'm pleased with the result, now this little babe sounds like a cranked Marshall, listen to 'Woman' from Wolfmother and there it is!

I like guitar, to experiment with sounds, alot.. You can't imagine all the designs I have in my little head, but I need to understand the maths first!

Indeed, there's no end :)

Edit: Right now I'm looking to isolate the cascode section of the firefly amp, to make a kinda 'preamp' or boost, I'd love that in front to certain of my amps,. I found this and you can't imagine right now how much questions jiggle in my head ;)

Edit2: I looked inside a 12AX7, I can't really say which filament is closer to his cathode, really you have good eyes! :D

Thanks guys!
 
Last edited:
Each filament is very close to it's cathode, 'cause it's sitting inside of it ;).
Decide which triode you'll chose as the 1st one right behind the input jack. Many tube data sheets say that this should be the one over pins 6 to 8. Next look to which pin bsiedes #9 it's heater is connected (I guess it's #5) and tie this pin to ground. Or, per Gnobuddy's suggestion, try swapping the heater leads to pins #4 and #5 in order to minimize hum.

Best regards!
 
Thanks a lot for sharing tips coming from long time experience!
I cannot claim long experience with valves/tubes - I only began tinkering with them around 2010 or so. I had lots of previous experience with solid-state audio electronics, though, which speeded up the valve learning curve.

where's the best spot in the chassis to ground the 12AX7?
I usually use a bare copper wire ground bus, and ground each gain stage to it very near the corresponding valve. I use a plastic (insulated) input jack, so the jack is not grounded at the front panel. The ground bus eventually connects to the first filter cap in the power supply, and that is the point at which I run one ground wire to the chassis.

I found that the A suffix in any tubes designates a controlled filament warm up time, I also found that it's supposed to be a lower noise version, is this true.
For what it's worth, Merlin Blencowe (aka the "Valve Wizard") says in his valve guitar preamp design book that suffixes on current-production 12AX7s are meaningless, and should be ignored.

Personally, I stay away from "tube rolling" and associated beliefs, some of which are clearly not based on reality. It's odd how the more expensive the valve is, the better it sounds to some people. :D

My $1 6JW8 must sound terrible according to that criterion, but luckily for me, it's the incredibly rare "purple plate, short-bottle, platinum-plated pin, NOS RCA" version, which was packed gently in unicorn-fluff lined jewelry boxes by princess-mermaids working the assembly line in 1945. So it's really worth millions of dollars, and therefore has fantastic sound, even though I only paid $1 for it. :nod:

I wonder if it worths to make a ground switch
To me, it wouldn't be worth it. You may find negligible difference in hum whichever way the heater wire is connected, who knows. Even if there is a difference, it will stay the same for the life of the valve, so there is no reason to ever reverse it once you find the quieter position. So my suggestion is, just tack-solder the 6.3V AC wires one way, then the other, see if there's any audible difference in hum.

listen to 'Woman' from Wolfmother
Interesting. Wolfmother really does sound like Led Zep merged with Deep Purple!

...I'm looking to isolate the cascode section of the firefly amp...
You can definitely do that, if you like. It is odd, though, the ECC83 / 12AX7 isn't normally a good valve to use in cascode, because each triode wants around 200 volts across itself, so the cascoded pair really isn't happy sharing a single 300V supply rail.

...I looked inside a 12AX7...
Sometimes you can learn surprising things by peering into a valve with a magnifying lens. I found two valves which are "pentodes" according to the data sheet - but when you look inside the actual valve, they are in fact beam tetrodes, with only two grids, not three!


-Gnobuddy
 
Okay!I found that the A suffix in any tubes designates a controlled filament warm up time, I also found that it's supposed to be a lower noise version, is this true.
It may (for some tube types), or it may not (for others). Have a look at the individual datasheet.
I'm sure that the original (NOS) 12AX7A had hum-supressing, bifilarly arranged heater filaments, which a 12AX7 hasn't. Hence the original 12AX7A equalled the European ECC83 and the 7025 equalled the E83CC, respectively.
Best regards!
 
Hey guys,

I thought the ground switch could be nice when swapping for another tube but if you say that it doesn't worth I'm gonna keep this in my knowledge pocket, maybe it will serve later :)

I usually use a bare copper wire ground bus, and ground each gain stage to it very near the corresponding valve. I use a plastic (insulated) input jack, so the jack is not grounded at the front panel. The ground bus eventually connects to the first filter cap in the power supply, and that is the point at which I run one ground wire to the chassis.

It makes sense I saw that often while looking at pics online, but I see so much designs I can't be sure.

One say this and that.. What do you think about star grounding?

Personally, I stay away from "tube rolling" and associated beliefs, some of which are clearly not based on reality. It's odd how the more expensive the valve is, the better it sounds to some people. :D

Neither I am, tube vendors are good at raising prices when a tube get popular, and some people have the money so.. That's why I love diy.

My $1 6JW8 must sound terrible according to that criterion, but luckily for me..

I have many (unknown to me) tubes, maybe one day I'm gonna be able tu use them! ;)

You can definitely do that, if you like. It is odd, though, the ECC83 / 12AX7 isn't normally a good valve to use in cascode, because each triode wants around 200 volts across itself, so the cascoded pair really isn't happy sharing a single 300V supply rail.

I think it has been said by the same guy, I mean Mr. valve wizard.. so far the amp sounds pretty good. Could it be better with a 12AU7?

I had a look at his cascode tutorial page and there's something I'm not sure, is the signal coming out the cascode inverted or not?

It may (for some tube types), or it may not (for others). Have a look at the individual datasheet.
I'm sure that the original (NOS) 12AX7A had hum-supressing, bifilarly arranged heater filaments, which a 12AX7 hasn't. Hence the original 12AX7A equalled the European ECC83 and the 7025 equalled the E83CC, respectively.
Best regards!

Okay! That's why they are more suitable in a daisychain design.

So now back at my little amp, and while we're speaking about datasheet I was wondering what would be the best OT for this amp.

I looked at the datasheet and found the 50L6 is looking for a 2000 ohms load resistance.

Most of my speaker cabs are 4 ohms

I did 2000 / 4 = 500

I get an impedance ratio of 500:1

If the impedance ratio is the square of the turns ratio, then the turns ratio is the square root of the impedance ratio.

Taking the square root of 500 gives me a 22.4:1 ratio

Now when I look on the internet (radio part stores) I see many OT from various ratios which are listed to be used with 50L6/50C5 tubes.

If for example I raise the load impedance on a tube by choosing an OT with 27:1 ratio, then use it with a 4 ohms speaker, the result will be less power output but less distortion, am I correct?

For now, I have some cleaning to do on my little workdesk.. be back later! :)

Thank you!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.