What with the current interest in tube projects on here I thought I'd share something I've been working on for the last month or so.
It's a simple low voltage tube amp using a couple of 12AT7s and a cheap transformer. I was originally going to build a Sopht amp (http://www.sophtamps.ca/mambo/index.php?option=com_content&task=view&id=25&Itemid=37) but the tubes didn't seem that easy to get hold of in the UK so I thought I'd try and make something using "normal" tubes.
After a bit of experimenting here's what I came up with.
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=36509&g2_serialNumber=2)
The preamp is basically just copied from the Fender Princeton and the output stage was inspired by the reverb driver on a Twin. I used a 100volt multitap line transformer (http://www.theelectronicsshop.co.uk/Infomation/Loudspeakers.htm) because I've read about similar things being used for Sopht amps and they're cheap. Mine has taps for 1.9W, 3.75W, 7.5W, 15W and 30W on the primary side and 8 and 16 ohm taps on the secondary. I'm using the 1.9W and 8 ohm taps. The transistor buffer between the stages is needed because I found I could only get the "power amp" to work with a very low input impedance. The tubes in the preamp can't drive this so I added the buffer, this is directly coupled to the second preamp tube's plate, which I copied from Doug Hammond's Chime and Grind amp (http://www.aronnelson.com/gallery/main.php/v/DougH/amp/chime/).
This means the signal path isn't 100% tube but if it's good enough for Doug it's good enough for me :)
Here's a quick soundclip (http://www.aronnelson.com/gallery/main.php/v/slackers-stuff/rascal/rascal.mp3.html). The beginning as about as clean as it gets and the last section is with the volume on max. I've no idea what the wattage is but through a 10" speaker it's loud enough as a bedroom practice amp.
I've still got a few issues to sort out, there's a bit of mains hum that seems to have become more noticeable since I boxed it up and I'm still working out the best way to provide power to the heaters.
Thre's a couple of pictures (http://www.aronnelson.com/gallery/main.php/v/slackers-stuff/rascal/) in my gallery if anyone's interested.
Any suggestions for improvements would be gratefully received.
Looks good Ian. ;)
How are you powering it? Do you think it would be possible to use something like a MAX1044 or LT1054S to get the plate voltage from 9 or 12 volt wall wart?
Looks and sounds great. Its got a nice crunch...it seems like it really responds to your feel/playing.
Did you build it on vero?
Have you tried dealing with the lead dress? Seperate the AC and DC wires and maybe shorten some wires up ???
Quote from: frequencycentral on September 16, 2008, 05:19:43 PM
Looks good Ian. ;)
How are you powering it? Do you think it would be possible to use something like a MAX1044 or LT1045S to get the plate voltage from 9 or 12 volt wall wart?
Rick, I was wondering if a charge pump could help us in our tube experiments, but i read you get very low mA, so I believe it wouldn't meet the higher mA requirements of the tubes...
Quote from: frequencycentral on September 16, 2008, 05:19:43 PM
Looks good Ian. ;)
How are you powering it? Do you think it would be possible to use something like a MAX1044 or LT1045S to get the plate voltage from 9 or 12 volt wall wart?
At the minute I'm using a 24volt AC power supply which is then rectified to get 36 volts. The heaters are run off 12.6 volts using an LM317 this is the bit I'm still playing about with because the regulator needs a fairly big heat sink.
I'm not sure about using voltage multiplier chips it might be possible to do something with one. I'll have to measure it but I think the LT1054 could handle the current draw of the circuit.
Quote from: kurtlives on September 16, 2008, 05:26:08 PM
Did you build it on vero?
Have you tried dealing with the lead dress? Seperate the AC and DC wires and maybe shorten some wires up ???
Yeah it's on vero. The cables are still a bit of a mess because I haven't figured out how the power is going to be wired up yet. Some better lead dress might help with the hum.
Quote from: Renegadrian on September 16, 2008, 05:40:37 PM
Quote from: frequencycentral on September 16, 2008, 05:19:43 PM
Looks good Ian. ;)
How are you powering it? Do you think it would be possible to use something like a MAX1044 or LT1045S to get the plate voltage from 9 or 12 volt wall wart?
Rick, I was wondering if a charge pump could help us in our tube experiments, but i read you get very low mA, so I believe it wouldn't meet the higher mA requirements of the tubes...
From memory the MAX1044 is 10ma output, maximum input voltage 10.5 volts. The LT1054S is 100ma output and will handle 12 volts input. I've used MAX1044 with tubes in some experiments. I think it's the heaters that require the high ma, not the plates. Just curious if it would also handle the transformer.
Looks and sounds cool Ian
The clip is a bit papery sounding but I'm sure it's just the laptop mic.
Does it hold it's own compared to a typical decent sounding amp?
(Not that it doesn't sound decent :icon_wink:
john
do you have a vero layout for the newbs like me who can't read schematics well? ::)
Quote from: Br4d13y on September 17, 2008, 01:00:38 AM
do you have a vero layout for the newbs like me who can't read schematics well? ::)
Ask about symbols that you don't know ! A lot of guys (and girls: I wish there would be some more...), here will inform you ( pleases don't be shy), and if other are condescendant in a bad way,
outline it. And after, you will be able to do more things: from the knowledge that your friends have given you. And if you have more questions, ask again... The more you will be specific, the more it will be profitable for you ( and also for the others) to be on this forum with us !!!
Yours !
F.H.P.
P.S. : Schematics reading is not so difficult ! There are
not a lot of symbols to know about what you have to solder, aside from different ICs specific to some boxes. Again, ask specific questions (or general ones, and mention (and outline them) what is obscure to you) !
Be
patient and kind, and you will get a lot of feedback (I assure you ! ) !
Slacker have posted a lot here (and his project is important to him, and I am sure that he is proud of it ) , and thus have contribued to the "friendship" athmosphere of this forum.
Welcome and be gentle here (and slowly learn), because everyone here want it this way.
A specific question is a contribution.
Yours again.
:) ;)
And be assured that you are welcomed ! :D
Again: Ian have posted something that is not common since everybody on WWW is searching for new guitar effects, and his work is a contribution to the forum and persons that had been here for a little while. Please cautious and kind !
PS.: Change your name: nobody will remember what it is... Even if it is not your "real" name... We don't care. ( Well, I don't, as long as I (and others) can remember you )
F.H.P.
( AND BE SURE NOT NO BE SCARED BY ANYBODY, AS EVERYONE WILL BE WILLING TO HELP YOU ! STAY A WHILE AND YOU WILL GET MANY ANSWERS FRIENDS !
Power stage: Is that class A in parallel?
Why not push-pull as in Firefly?
Quote from: John Lyons on September 16, 2008, 07:25:49 PM
Looks and sounds cool Ian
The clip is a bit papery sounding but I'm sure it's just the laptop mic.
Does it hold it's own compared to a typical decent sounding amp?
Thanks John. In real life it sounds fuller than in the clip, the laptop mic loses some of the bottom end. The speaker used wasn't in a cab either it was just lying on a box, so it's probably not the best advert for the sound.
I've since tried it through one of my Twin's speakers and it sounds a lot warmer and smoother. I'll get hold of a mic and do some better soundclips.
I don't know how it would stand up next to a proper low wattage tube amp but as an experiment I'm more than happy with the results.
Quote from: Steben on September 17, 2008, 03:17:54 AM
Power stage: Is that class A in parallel?
Why not push-pull as in Firefly?
Yes it's single ended class A with 2 triodes in parallel to get a reasonable volume. Push pull would probably be better but I think I'd need a proper output transformer for that and I wanted to do this on the cheap. Now I know the idea works I might try a push pull version.
"The speaker used wasn't in a cab either it was just lying on a box,"
In that case the clip sounds incredible. :-)
cheers.
Slacker, have you wired the heaters on each tube reverse phase to each other? The other trick is to have a variable resistor between the pins to adjust the midpoint. I've picked them up in old valve tape machines but have yet to find a 200ohm wound pot. Another trick with heaters on AC is to do the centre tap with 100 ohm resistors then attach the mid point to a reasonable high DC point at high impedance. You could try about 24V.
Oh yeah, and the sound is noticably 12AT7! I used to be able to get old Tesa brand 12AT7's for a few bucks each and made loads of things with them.
Oops, I see now that you're running your heaters on DC so none of the AC tricks would apply. But from your photo it looks like your power in is those twisted pairs and the heater pair, are they the grey and orange?, aren't twisted anywhere near tight enough and they could be set alot closer to the chassis and away from the other wires. The best trick I got for grounding tube amps was that all the grounds should spider to the highest impedance point and that was always the input.
Given your power supply is coming in as DC then the grounding issues are whats bringing in AC 60Hz noise 'cause even ripple is at 120Hz so your grounds must be an antenna to bring in 60Hz... ??? unless it's coupling into the DC lines in. A wooden chopstick to push twisted pairs about is the best tool on p-p amps.
Cheers for the advice Sean, I've still got to sort out where I'm going to mount the rectifier and regulator so once I've done that the wiring should hopefully get a bit neater. I'll definitely try increasing the amount of twist in the power and heater wires and route them as out of the way as possible.
I think I just need some extra filtering on the power supply to get rid of the hum, but I'll have a better look at it at the weekend.
I'm the junior assembly guy at Kendrick Amplifiers, I don't have a proper college education and I'm constantly learning, so take this advice with a grain of salt.
One thing to keep out noise/hum is you could try mounting less on that board and more on the panel mounted components. Instead of having signal go from a pot lug to a wire to a cap on the board to another wire to another pot lug, just mount the cap from lug of one pot to the lug of the next, you'd eliminate about a foot of wire (antenna wanting to pick up noise). Also, make sure you're not daisy chaining your grounds, connect each ground point to something that is directly connected to your chassis which is nearby (sleeve of a jack, back of a pot, grounding lug on one of the bolts holding in your transformer or tube socket), this eliminates the chance of ground loops within your circuit.
Anyways, congrats on the build and good luck with tweaking it out and getting it as proper as you want it to be. From your photos, I couldn't tell exactly what was going on all the time, so if some of what I wrote above was something you were already doing, sorry.
Oh and also, I agree with twisting those wires into tighter pairs.
Yeah, tubes are such a completely different kettle of fish compared to transistors. High Voltage at high impedance sharing the exact same space as Low Voltage at a low impedance so the territory becomes rife for all sorts of magnetic coupling to occur.
I absolutely love point to point wiring on tube amps and have to admit that when an amp works quietly re; S/N, it's usually quite beautiful, aesthetically, as well. Tubes are mystical, and getting them to work right makes life alot easier to live as if being able to balance those two extreme variables, HighV and Imp and LowV and Imp, somehow affects the psyche and works on ones enviro and Inviroment...
Sorry Chaps... Tubes are like that for me!
Quote from: Boogdish on September 18, 2008, 05:29:02 PM
I'm the junior assembly guy at Kendrick Amplifiers, I don't have a proper college education and I'm constantly learning, so take this advice with a grain of salt.
One thing to keep out noise/hum is you could try mounting less on that board and more on the panel mounted components. Instead of having signal go from a pot lug to a wire to a cap on the board to another wire to another pot lug, just mount the cap from lug of one pot to the lug of the next, you'd eliminate about a foot of wire (antenna wanting to pick up noise). Also, make sure you're not daisy chaining your grounds, connect each ground point to something that is directly connected to your chassis which is nearby (sleeve of a jack, back of a pot, grounding lug on one of the bolts holding in your transformer or tube socket), this eliminates the chance of ground loops within your circuit.
Anyways, congrats on the build and good luck with tweaking it out and getting it as proper as you want it to be. From your photos, I couldn't tell exactly what was going on all the time, so if some of what I wrote above was something you were already doing, sorry.
Oh and also, I agree with twisting those wires into tighter pairs.
If there is one thing I have learned from Gerald it's this. ;D
Quote from: frank_p on September 17, 2008, 01:45:24 AM
PS.: Change your name: nobody will remember what it is... Even if it is not your "real" name... We don't care. ( Well, I don't, as long as I (and others) can remember you )
I think his name is Bradley, only some letters are transposed to numbers. I'll sure remember him now that I cracked the code! :icon_cool:
Greetings
Miguel
Quote from: bioroids on September 18, 2008, 11:19:58 PM
Quote from: frank_p on September 17, 2008, 01:45:24 AM
PS.: Change your name: nobody will remember what it is... Even if it is not your "real" name... We don't care. ( Well, I don't, as long as I (and others) can remember you )
I think his name is Bradley, only some letters are transposed to numbers. I'll sure remember him now that I cracked the code! :icon_cool:
Greetings
Miguel
Miguel, I feel stupid to write long messages like this, just for someone asking for a vero. I should have kept my hands off the keyboard... This was not my business, but hey, Ian is still perfecting his creation...
Anyway, sorry for the OT, Miguel "Cracker-Jack" !
Ian, did you ever look at this? http://www.jjs.at/electronic/class_a_subminiature.html
The designer states that his NE555 based power supply isn't exactly safe, but I would love to know if using a LT1054 would be sufficient to power a small tube amp from a 12 volt wall wart.
Yeah I had seen that before, I didn't realise or didn't remember that he'd also gone with a Fender reverb type arrangement though, just goes to show that nothing's ever really a new idea I guess.
I like the idea of getting proper tube voltages out of low voltage wallwarts but I for now I'm happier sticking to low voltages.
I'm pretty sure the LT1054 would work fine. I've taken some measurements and the amp its self (discounting the heaters) only draws about 10ma at about 33 volts. It also works at lower voltages, I've tested it at 20 volts and it draws about 6ma, it sounds Ok at this voltage but is a bit quieter than at 33 volts, any less and it distorts too much and in my opinion is really too quiet to be of any use.
I've only had a quick look at the datasheet so I might have missed something but the LT1054 is rated for input voltages up to 15volts and the basic positive doubler circuit can provide 50ma. Taking into account the losses mentioned in the datasheet you should be able to get about 22volts at 50ma from a 12 volt wallwart, this is more than enough for the amp. Next time I need some parts I'll order a couple and have a play with them.
Cheers to everyone for the tips on getting rid of the hum. I've solved it by adding an additional RC filter between the rectified AC and the amp.
Maybe try a power mosfet at the output as opposed to a NPN Transistor...The mosfet would probably give you more output current ,Probably breakup more symetricly and with an output cap you could drop the output transformer...With a strong power supply you might be able to get 10w with 36v Rails....Just a thought....
Cheers
I think if you set up the second triode as a cathode follower you could forgo the transistor all together. Although, to do that the two have to be DC coupled (no decoupling cap) so that would change your tone stack placement, and putting it after the follower may make doing so useless, not providing enough current to drive the power amp. If you added a second tube you could do this immediately following and coupled to the second triode, or you could borrow a triode from the power amp, taking them out of parallel.
Maybe none of this meets your design goals, but a bit of information for others perhaps. To make a cathode follower after stage two, output from the anode of triode 2 directly into the grid of triode three. Triode three's anode should be connected directly to your B+ and the cathode should be connected to ground through a resistor between 47k and 100k (lower for more compression on the negative side of the waveform (I think). Take the output from the cathode this time (between the cathode and your resistor) through a decoupling capacitor into your power amp.
I guess maybe I should draw that up.
Quote from: frequencycentral on September 21, 2008, 06:09:19 AM
Ian, did you ever look at this? http://www.jjs.at/electronic/class_a_subminiature.html
The designer states that his NE555 based power supply isn't exactly safe, but I would love to know if using a LT1054 would be sufficient to power a small tube amp from a 12 volt wall wart.
oh my gosh! did anybody not notice this looks
exactly the same as a zvex nano head? hmmmmmmmm
Quote from: Br4d13y on September 24, 2008, 12:44:46 AM
Quote from: frequencycentral on September 21, 2008, 06:09:19 AM
Ian, did you ever look at this? http://www.jjs.at/electronic/class_a_subminiature.html
The designer states that his NE555 based power supply isn't exactly safe, but I would love to know if using a LT1054 would be sufficient to power a small tube amp from a 12 volt wall wart.
oh my gosh! did anybody not notice this looks exactly the same as a zvex nano head? hmmmmmmmm
oh course, there was a thread about it
its not the same circuit though
oh. ;D
Here's a much better clip of the amp http://www.aronnelson.com/gallery/main.php/v/slackers-stuff/rascal/rascal2.mp3.html (http://www.aronnelson.com/gallery/main.php/v/slackers-stuff/rascal/rascal2.mp3.html)
This was recorded using a real mic and 1x10 closed back cab. The clip starts off clean, then with volume at about 75% and finally on full. Then at the end I'm boosting the already dirty amp with a Colorsound Overdriver.
I think that I like the Clean to slightly broken up sounds best.
The clean sound has a nice sparkle to it.
The overdriven sounds are a little harsh sounding to my ear.
Thanks for the new clip.
john
Yeah I'd agree with that John the lower gain settings work best. The higher gain settings are a bit rough and the bottom end gets a bit flabby, I don't know if it can be tweaked to sound better or whether I've reached the limits of what I can do with it.
The coupling caps are pretty small as is so there should not be too much bass making it sloppy down there.
Mainly I think it sounds a little grating for the higher gain settings.
Something to smooth out and gain some clarity and note definition when the volume is up would make it better.
I'll have to look at it some more.
john
Hi. I'm going to try your amp and was thinking in using 4 x 9V batteries for the 36V plate and a lead acid battery for the heaters. That would make a lovely battery powered amp! (...and eliminate AC induced hum) A 2.2 Ah battery would last for more than 4 hours easy and the 9V should last about 12 hours.
I like the crunch sound but I really love the cleans of it. Making the amp cleaner would be great so one could always overload it externally (with a small BS170 booster (*) for example or your OCD, etc.)
No idea of how to make this amp cleaner without lowering the overall power...
Some ignorant intuitions :
- A more sucking tone-stack (Ampeg baxandall? Fender AA764?) may make it cleaner but lower the power too...
- Using 12AU7 for the pre maybe...
- Adding a master vol control before the 12AT7 "amp" to keep it at max power before saturation when you like? (don't know how to do it)
- supress the 33uF bypass cap?
- add negative feedback?
Any clue?
Fernando
(*) http://www.kilback.net/homebrewtweaks/pedals/cleanboost.htm
I think your idea for running it on batteries will be fine, it's only the heaters that need a lot of current.
I tried adding negative feedback and that does clean it up a bit at lower volumes, it's still dirty at maximum volume though. Using a 12AU7 for the power tube makes it cleaner and adds a bit of sparkle, it sounds good but I prefer the grit of the 12AT7.
A 12AU7 in the preamp sounds brighter and has bit of compression in the midrange, I'm not sure if it's any cleaner though.
I haven't tried a master volume so I'm not sure how well it would work.
hi, thanks a lot for the help
Don' you think the transistor may be too easely overloaded? Maybe a vol control just before the transistor would permit to avoid that whenever you like... so you get only tube saturation.
Wich would be the correct way of controling the amplitude of the signal going to the transistor?
duplicate, sorry
any solution found to eliminate the transistor?
comfortably_numb did you draw your suggestion?
>>>
I think if you set up the second triode as a cathode follower you could forgo the transistor all together. Although, to do that the two have to be DC coupled
(...)
I guess maybe I should draw that up.
>>>
Hi
a couple of suggestions, which may not be what you're looking for:
1. Power supply. What about a printer's supply? My $49 injet printer (now busted) came with a supply that gives 32V AC 400mA and 16V AC 600mA. You could probably pick up such a thing for $5 at a charity shop in the electronics section (mine sells wall-warts for 50c).
2.
QuoteYes it's single ended class A with 2 triodes in parallel to get a reasonable volume. Push pull would probably be better but I think I'd need a proper output transformer for that and I wanted to do this on the cheap. Now I know the idea works I might try a push pull version.
You don't need a push-pull transformer at all. Just use two taps that are equally spaced (in either DC or AC ohms) from the centre tap. I'm a little surprised that the low impedance primary works so well for you. That puts a low load on the output valve (which the AU7 should handle better than the AT7). If you do decide to go to push-pull, I think you'll find that you can get around 0.2W from such a system. (ie almost as much as a Gem/Ruby/Noisey cricket)
nice work!
Just listened to your "better" soundclip. Sounds great!
Quote from: brett on October 27, 2008, 07:46:36 AM
1. Power supply. What about a printer's supply? My $49 injet printer (now busted) came with a supply that gives 32V AC 400mA and 16V AC 600mA. You could probably pick up such a thing for $5 at a charity shop in the electronics section (mine sells wall-warts for 50c).
That's a great idea Brett, I'll have to see what I can find locally.
Quote
You don't need a push-pull transformer at all. Just use two taps that are equally spaced (in either DC or AC ohms) from the centre tap.
Thanks for the info, I'll try a push pull version and see what happens. I found this site (http://ozvalveamps.elands.com/optrans.htm) that has some info about using these transformers with real valve amps. I'm trying to digest it all at the minute, but it looks like useful stuff.
Quote
I'm a little surprised that the low impedance primary works so well for you. That puts a low load on the output valve (which the AU7 should handle better than the AT7). If you do decide to go to push-pull, I think you'll find that you can get around 0.2W from such a system. (ie almost as much as a Gem/Ruby/Noisey cricket)
I'll be honest I was surprised it worked at all! For the output section I basically just built something that looked like an amp with out any real technical knowledge and it worked, guess I just got lucky.
Quote from: DougH on October 27, 2008, 11:42:39 AM
Just listened to your "better" soundclip. Sounds great!
Thanks Doug.
wich would be the ideal primary impedance then? (of the OT)