Lightweight guitar amp ideas... Now with compactron! (and schematic, more ?s)

[soapfoot]

I'm thinking of designing a guitar amp with the goal of keeping weight to a minimum.  I'm less concerned with maximizing power-per-dollar than I am with maximizing power-per-ounce.  I was kicking around the idea of going single-ended, and I'm wondering if this might be a good way to keep weight down.  My thinking was that fewer tubes drawing filament current would allow a smaller, lighter power transformer, and I assumed that a single-ended OT could be smaller/lighter as well (that might be incorrect).  Not to mention smaller chassis (one fewer tube socket), etc.

If you agree that this might be a good way to go to achieve my objective, the question becomes "which power tube to use?" An obvious choice would be a 6550/KT88, but if something was capable of putting out more power in single-ended operation, I'd be willing to spend more, hunt harder, etc.

I want to drive a single efficient 12" speaker, and the idea would be to keep weight down as much as possible while still being able to drive the 12 to moderate volumes. I already have weight-saving ideas for the speaker (neodymium) and cabinet (paulownia wood), and I'm willing to suffer the expense of getting one-off transformers made.  I do want the amp to sound great in a semi-'traditional' way as a guitar amp, though, and as a result I'm reluctant to use toroidal transformers for either power or output.  By using a bridge rectifier, I could get by with using only two tubes in the amp.

The goal would be to make a great-sounding 1x12 tube amp, usable on gigs, that weighs 15 lbs or less.  Any ideas or direction is welcome.

 

[soapfoot]

Been doing some research, and it seems that New Sensor is making something called the "KT-120."  It claims 60w plate dissipation, which is quite an increase over the KT88's 35w.

Here is the data sheet: http://www.newsensor.com/releases/KT120TungSol.pdf

curves: http://www.newsensor.com/releases/KT120TungSol-Curves.pdf

I think this might be a winner.  It says maximum V(a) is a staggering 850v DC!  Would I go with something that high, or do something more sane like 600v?

 

[Kingston]

With that tube you've shot out of the "lightweight" range. Someone needs to deliver the current for all those watts. And that someone is a very heavy and large sized PSU transformer. There are significant heater current needs as well.

And then someone needs to deliver those watts to the speaker. And a transformer handling even 50 watts of single ended power is a big boy.

When I originally looked at your specs I thought "parallel EL84 SE output with 12AX7/12AU7 input stages and filtering". Not many watts in that, but can be kept considerably small and would sound great. Needs only a PSU transformer and the SE output, perhaps upgrade with a PSU choke and a tube rectifier. I would first start with something like that. Many existing designs to be adapted as well.

 

[soapfoot]

thanks for your thoughts.

Still brainstorming, I wonder if there's a safe way to do the old "series heater" trick without the electrocution risk. Might not get the sound I want, but it's an idea at least, and the lack of a power transformer would certainly reduce weight.

If I was going to do two EL84, what would be the advantage of parallel single-ended versus going ahead and doing a class A/B design?

 

[Kingston]

what would be the advantage of parallel single-ended versus going ahead and doing a class A/B design?

Technically there are no advantages. Class A/B wins in every respect, usually less distortion, always more watts.

Expect perhaps SE would sound better? That's debatable, but I've always preferred the fender champ style distortion to its class A/B counterparts like blues junior or pro junior. But a very hot and distorted class A/B has produced some seriously good sounding tunes as well.

In the end it comes down to taste.

[edit]

I made a mistake above. Originally wrote tweed when I meant champ. That one is SE. I get confused with fender names sometimes, as they so many revisions of everything.

 

[gemini86]

A high output se ot will have quite a gap for DC current. Then to keep the inductance up, it will need more iron. If you're shooting for lightweight then power tubes aren't a viable option. You might be able to use a amps for heater power and that could cut weight quite a bit, but there's no getting around that huge ot unless you series a few 32 ohm speakers and go otl output.

 

[lassoharp]

I don't know exact weight of one but I would first ask if Fender Champ is the right size for you.  I'm guessing it would be since you want 12" speaker.

I have a SE amp using 6L6 for output tube - the OT is very small,  sits on a 5 X 10 chassis and will raise hell pretty good into a 4 X 10 cab.  Into a smaller single speaker with open back cab it's a different animal in terms of bass response and bigness of sound.  Excellent tone either way.  I doubt it puts out more than 6W. 

6L6 may be slightly more economical than 6550.  Won't hear the difference in a few watts.  I would assume you're going to mic amp on stage? 

The big sound comes with a big cabinet.  When I think of SE I think of tone as the main reason people go there.  PP amps have to be pushed into "Bloom" before they get fat like the SE is at all volumes.  Even good Blackface Pros & Twins sound comparatively hard and thin at low volumes

The Hammond SE OT and PT can be had at reasonable cost and are great quality.  Edcor and Weber have also have both.  Bigger cost may be the cab & chassis.

I'd make speaker choice based on main musical style. 

The goal would be to make a great-sounding 1x12 tube amp

Champ with 6L6, EL34, or 6550 output + 12in speaker (old celestion 15W if you can find one)

 

[PRR]

The SE OT will weigh 2 to 3 times as much as a p-p OT of the same power.

I know this is a DIY forum, but.... the low-cost tube amps on the market are probably as LIGHT as it gets for the power. Iron costs money. These guys musta sharpened their pencils, and they have access to custom transformer production.

Peavey ValveKing 112 Combo Amp -- 50W $400 44 lb
Bugera V55 55W 1x12 Tube Guitar Combo Amp -- 55W $500
Peavey ValveKing 100 Head -- 100W $550 45 lb
Peavey ValveKing 212 Combo Amp -- 100W $650 69 lb
Blackstar Venue Series HT Stage HT-60 Combo Amp -- 60W $900 67 lb

These things will surely be frilled-up copies of the Marshall which is based on the Fender 5F6-A. The topology is good. The exact values in the tone and distortion networks may be extreme to give a "WOW!" sound which cuts through a showroom. Going back to mostly Marshall/Fender part values in the main signal paths will probably make a more useful amplifier.

Peaveys are usually robust, reliable, and as easy to service (or modify) as any mass-produced PCB product can be. And worst-worst, you toss the PCB, keep the chassis and iron, wire-up a 5F6-A on turret board. I know you can't DIY a 100W amp for anything like $550, and 45 lb seems light to me.

> a safe way to do the old "series heater" trick without the electrocution risk

No, not for modern expectations of "safe". And with 6L6 (what I would use for "light" in the 30W-150W zone) you would not save much. 100W of 6L6 is 200W of B+ and 23W of heater. Yes, less, but not much, and heaters DO short to cathodes (and axe strings).

> what would be the advantage of parallel single-ended versus going ahead and doing a class A/B design

For the same output power: SE keeps you warmer in winter.

 

[Jarno]

The big Peavey full-tube bass head uses a SMPS, not sure if the smaller guitar amps also use a switchmode supply, but this would shave off some weight.
Using an Alnico speaker too.

 

[soapfoot]

I already have plans for most of what I'm going to do to reduce weight.

First off, I'm not going AlNiCo... I'm going Neodymium.  The Jensen Jet Toronado is an efficient 12" speaker that weighs only 4.45 lbs. 

Secondly, my chassis will be aluminum instead of steel

Finally, I'm going to make a cabinet out of Paulownia. If you don't know about this wood... it has strength comparable to poplar and weight only slightly more than balsa.  About half the weight of pine (typical Fender combo amp wood) with similar sonic characteristics (when used in guitar bodies), I will further weight-relieve it by thinning the panels a bit. I'm even considering some 'blind' holes (that do not go all the way through) on some of the panels for further weight relief. Since my speaker will only weigh 4.45 lbs, I can get away with a 3/8" ply baffle, instead of the customary 1/2" or 3/4".

The cabinet will compromise durability a bit, but this is an acceptable tradeoff for me.  I have plenty of more durable amps for more rigorous situations. This one is going to be made specifically for those occasions where I have to carry an amp on the subway from my house 7 blocks away.

I will also install two handles on the sides rather than one on the top.  Carrying the amp in this manner, even when light, feels infinitely more natural to me than the imbalance of having one weight off to one side. Much better on the back to be balanced. Could still be carried with one hand if need be.

As for the design of the amp, here's what I've learned thus far--

high-powered SE is out due to increased OT weight.  This is contrary to what I expected.

However, I'd still like to keep the amp down to 3 bottles so I could use a Champ-sized aluminum chassis.  This has me thinking...

A friend mentioned the 6U8, which is a pentode and a triode in the same bottle.  Perhaps I could use the pentode for a preamp stage, the triode for a cathodyne PI, and then a pair of 6L6?

Then there are tubes like the ELL80, which is a dual output pentode in one bottle.  It's only good for about 6-8W in push-pull, but perhaps someone knows of a similar tube that can give a bit more output?

One more question-- would a 16 or even 32 ohm speaker allow for a smaller OT?

 

[Jarno]

Oh yeah, Neo wil be even lighter, excellent.

You could even go glass/carbon fiber epoxy or polyester construction, that'll be even lighter, I think you would have to "do stuff" to ensure damping of the panels or they'll ring forever.

And, ehm, perhaps single ended mosfet poweramp and tube preamp? (dives for cover)

 

[gemini86]

I was thinking he should go chip amp output... I'll take my tomato pelting now.

 

[lassoharp]

high-powered SE is out due to increased OT weight.  This is contrary to what I expected.

http://www.hammondmfg.com/125SE.htm


http://www.edcorusa.com/category/69-gxseseries.aspx    (click on details for weight)

A friend mentioned the 6U8, which is a pentode and a triode in the same bottle.  Perhaps I could use the pentode for a preamp stage, the triode for a cathodyne PI, and then a pair of 6L6?

The PT weight will always be considerably larger than the OT weight.  Going PP to save on OT weight with pair of 6L6 means a PT with roughly double the current capacity . .


The Paulownia wood sounds like a good idea.  I've been wanting to make a cab out of this for a while.  I've played through one of very thin construction and much deeper than standard Fender cab.  It gave solid body electrics a sound very similar to a hollow body guitar.  Very cool effect.

 

[soapfoot]

First and foremost, this is going to have to be an amp whose sound excites me.  If it's a lightweight amp that stays at home because I'm not in love with the sound, then it isn't serving its purpose.  I'm quite a traditionalist when it comes to the meat and potatoes of a guitar amp.  I just couldn't bring myself to do output transformerless or hybrid topology.  I've never tried a hybrid amp that I liked, and I'm not cocky enough to think I could make the first.

The Hammond OT in the link above doesn't interest me much because I don't like the idea of a rolloff at 100 cycles.  I will probably get transformers from Heyboer or someone like that, and I don't mind the expense of custom-wound iron if necessary.

However, as stated, if I could do a high-powered SE design and use a lighter PT, then I wouldn't so much mind a larger OT.  Going single-ended would also save weight in other ways, since I could likely do it with just 2 tubes.

Besides going toroid, are there any ways to reduce weight in a transformer?  Are there any lighter-weight magnetic materials that are commonly used for laminations (besides steel)?

 

[PRR]

1) Maximum useful efficiency of SE tube is 40%.
Maximum useful efficiency of PP tube (in portable sizes) is 55%.

2) SE operation rarely uses all the cathode current your heater-power "paid for". Peak current is small.
P-P operation can run considerably higher peak current, better use of heater-power.
One 6V6: 5 Watts; two 6V6: 14-20W.

3) (already mentioned) P-P needs much less iron.

> I don't like the idea of a rolloff at 100 cycles.

1) Guitar only goes to 82Hz; if it were really a "100Hz" -3db rolloff then 80Hz is only like 4db down.
2) The #1 way to save weight is to shave bass. Less cone-excursion (smaller magnet), less OT iron, less box.
3) If you gig with a bassist, the guitar "usually" (ha-ha) should rarely go into the lowest octave (80-160) or only as buzz-tone not true bass.
4) The Hammond 125SE series (already uninteresting because single-ended) has "simplified" specs. 125ESE is -3dB at 150Hz working as 10K load, but 75Hz as 5K load, and lower if you work 2K5 load (but this gets to be a big-heater bottle).

> would a 16 or even 32 ohm speaker allow for a smaller OT?

No. 1 ohm to 1K all the same. More/less turns of less/more fat wire. (Below an ohm, wire fatness may need more space; over some K the insulation adds space.)

> tubes like the ELL80

Way too expensive.

6DZ7 6DY7 -- note that these twin-tubes use more heater power than two 6V6 (albeit with modestly greater sensitivity), the DZ is $30 (not bad) and the DY is hard to find.

> use the pentode for a preamp stage, the triode for a cathodyne PI, and then a pair of 6L6?

Where is your volume control?

Power tube grids need around 20V; guitar amps are expected to make full power with 20mV from the jack. You need gain near 1000. Single pentode with practical impedance give gain 70-140. Two stages of 12AX7 give gain nearer 2500.

Study the classics. Single-pentode for all the gain is the first Champ. It hardly has enough gain, overloads easy. Twin-triode Champs are much better machines. But SE. In push-pull, the only one-small-bottle design I can cite is a Fender MusicMaster Bass which uses a driver transformer (and was not well-loved in its day). Pre-war Gibsons used similar but that was two single-triodes.

> a cathodyne PI

Cathodyne is essentially perfect. A high-volt transistor can be an equally perfect cathodyne with savings of sockets and heater power.

Alternatively: the first stage never hits gross overload, is not "tubey". A JFET with gain like 25 can be a first stage, allowing one pentode for second stage gain and flavor, pentode/triode for gain and PI.

But be wary of getting off the beaten path. Tweaking an odd-lot of unconventional tubes into a musical instrument can be endless. Plagiarize the past, you get very-close first try.

You have not, even roughly, stated your power-output goal. ELL80 and 2-6L6 are very different levels of "authority".

 

[soapfoot]

Thanks for the thoughts.  On the larger points, I agree with the philosophy of not straying too far from the classics (for many reasons, from pragmatic to sonic).

As for power requirements, here's some background. I make my living mostly as a guitar player.  Some of the bigger gigs/shows I get called to do are at venues with backline, but sometimes, for smaller things (or things in odd locations like radio appearances, restaurant gigs, etc). I do have to bring a small amp.  My current "subway amp" is a Pro Junior that I modified a bit and stuck in a slightly larger pine cab with an AlNiCo speaker.  It weighs 22 lbs and delivers 15-ish watts with two EL84 in p-p, driving a moderately sensitive 10" AlNiCo speaker.  This amp is about perfect in volume, and it sounds great.  It's almost light enough.  I can do the subway commute with it without too much dread, but I get tired after about 7-8 blocks with a guitar on my back.  If I have to carry a pedalboard, too, it starts to become a bit of a drag.  I hate rolling carts because I'm a somewhat tall person, and none of the handles on any of the folding carts i've found are long enough to make them travel easily for a person of my height.

My goal is to get an amp that might have a similar amount of clean headroom (a touch more would be welcome) but weigh about 5-10 lbs less, if that would be possible.  A 12" speaker will be more sensitive, and going neodymium won't increase weight at all over the 10" AlNiCo.  I want stone-simple... only a single volume control, one input jack, and an on/off/standby switch.

I'd say anywhere between 10-20 watts would probably be enough for what I need.  I'm flexible.  More power would be fine.

 

[soapfoot]

OK how about this....

6C10 compactron triple hi mu triode.  Two stages of gain, volume control in between.  Cathodyne PI driving a 6DZ7, giving probably 15-ish watts in push-pull from only 2 bottles.

What do you think?

 

[soapfoot]

OK, so here's my first draft of an amp with a 6C10 compactron and a 6DZ7.

I took inspiration from the black panel Fender Princeton and Deluxe, the Fender Pro Junior, and the Vox AC15 (just around the output section).

Anyone see anything wrong or have any ideas?

 

[soapfoot]

Here's a version 1.1 with a few minor corrections. I realized the screens of both halves of the 6DZ7 are tied together in the tube, and that I had the filter cap for my bias supply in backwards.

Would it be a good or bad idea to use one resistor and bypass cap to bias the first two stages?  Say an 820R bypassed with a 25µf, tied to the cathodes of V1A and V1B?

Now I have a few questions, mostly related to spec'ing a power transformer...

Since the 6DZ7 is like a pair of 7189s, what is the ballpark bias voltage they would need to see, assuming a plate voltage of 390v?  How would I calculate this?  In other words, I need to figure out what the transformer C- tap would be.

How much safety margin should I allow when calculating heater current capacity?  Current draw from a 6.3V bayonet base lamp, plus a 6C10 plus a 6DZ7, plus what percentage of that total for safety?  50%?  100%?

What specs do I use on the tube data sheets to determine the current handling capacity needed for the high voltage winding of the transformer?

Thanks for any help.

 

[lassoharp]

How much safety margin should I allow when calculating heater current capacity?  Current draw from a 6.3V bayonet base lamp, plus a 6C10 plus a 6DZ7, plus what percentage of that total for safety?  50%?  100%?

AC are heaters normally ran at full rated load so that final voltage will be as close to spec as possible.
You will likely run a little higher than 6.3V @ 50% and this will probably differ slightly from transformer to transformer.  If winding is rated 6.3V @ 2A, run at 2A, you will be fine.

The HT winding is often ran at 50% or less just so the transformer runs cool.