• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

DIY Tube Amp Questions

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
Hey all, I first should say I know nothing about tube amps.

Borrowed an amp from a friend who could not get it to make sound. He got it from a Electrical Engineer friend who built it and gave it to him as a gift. Anyway, being the engineer that I am (Mechanical) I could resist a challenge, especially on a neat peice of equipment like this, so I lugged it all the way home (weighs about 45 - 55 lbs)

It's a very nice looking, shiny SS chassis, tube amp that has (4) 6L6GC marked RUBY (I assume POWER Tubes), and (2) 6SL7 marked GE (assume TRANS Tubes), and (2) 5R4GYB marked RCA (assume RECTIFIER Tubes)

After some time, (missing & wrong fuses) I finally got it to produce sound. All tubes had a nice orange glow. Very nice with both my DIY speakers and my SMGa Maggies. Very smooth and dynamic, similar to my Panny SA-XR45 (I don't want to get into any discussions about Digital vs SS vs Tubes, so let it be please)


Question:

1.)It just doesn't sound very loud...now I know the maggies are a low impedance speaker and I shouldn't expect them to be very loud, but even my DIY speakers that are a good 92 db/watt were not very loud. I should say they were a good listening level, but not very loud. It was about 15% of the volume of my 100 watt XR45 level. Again comfortable, just not loud. Preamp was as high as possible.

2.) With this config of 8 tubes, anyone know approx how many watts/channel this would likely be? My friend says he thinks is 40 watts/channel. I would think a 40 watt/channel tube amp would be a bit more loud.

I opened the chassis and notices one of the three exterior huge CAPACITORS wasn't even connected - marked Mallory 20000 MFD. Don't know if the amp was going to be modified or what.

Any advice. My friend wants to sell it to me. So far, I am not so interested.

Thanks
 
After some time, (missing & wrong fuses) I finally got it to produce sound.
Good Show!
tube amp that has (4) 6L6GC marked RUBY (I assume POWER Tubes), and (2) 6SL7 marked GE (assume TRANS Tubes), and (2) 5R4GYB marked RCA (assume RECTIFIER Tubes)
Advice, tell your friend its useless to you without a schematic.
I don't think there is a circuit in which a 6sl7 could adequately drive a 6L6GC. However, Without seeing the circuit, I am making general statements.

With this config of 8 tubes, anyone know approx how many watts/channel this would likely be? My friend says he thinks is 40 watts/channel. I would think a 40 watt/channel tube amp would be a bit more loud.
The rule of thumb is a push-pull tube amp running class AB will output about 1/2 the combined plate dissipation of the output tubes. So a WAG would be 25 to 35 watts. If its running in Triode mode, that drops to about 1/4 the plate dissipation or 12-15 watts. That’s assuming well designed circuits.

I love tube amps. I would guess that this has potential, but would require a lot of work on your part.

Good Luck, Hope this helps.

Doug
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
> a Electrical Engineer friend who built it

Old guy? Or some new EE? Not all old EEs could build a 6L6 amp; but some of the tube stuff "built" by new EEs is awful ignorant.

> and gave it to him as a gift.

A "good" gift? Or "here, you can play with this" gift?

i.e.: is it supposed to be a good working amp?

The blown/wrong fuses and a disconnected cap of a VERY odd size sound confused.

> 6SL7 marked GE (assume TRANS Tubes)

I have no idea what "TRANS Tubes" are in this context.

Basics:

Do you know that tube amps can KILL? Can you work inside them so carefully that you cheat death? In Mechanical Engineer terms: it is a punch-press or a rotary-saw, except you can't see when the dangerous parts are dangerous.

Do you have 6.0 to 6.6V on the heaters?

Do you have +300VDC to +500V on the 6L6 plates? And a similar or somewhat lower voltage on the Screens?

Is the power transformer big enough?

You don't even say it has output transformers. If some "clever" EE has built an OTL 6L6 amp, no it is never going to be loud.

> just not loud. Preamp was as high as possible.

A proper tube amp should distort with a typical preamp maxed-out. It won't be the brick-wall clipping of a transistor amp, but it will clearly be in distress. And dual 6L6 per channel should be 20W to 50W, which should be loud enough.

It would be good to know the schematic. There are an infinite number of ways an EE could do it wrong. The simplest way to do it right would be 6SL7 voltage amp, 6SL7 Cathodyne. The first stage would give a voltage gain of 40. Cathodyne has unity gain to each 6L6 output. The outputs need 20V to 50V peak drive. So the raw amplifier needs 0.5V-1.2V peak, 0.3V-0.9V RMS, plenty of sensitivity. However 6L6 are not always considered hi-fi without negative feedback. Some EEs even understand NFB. Taking 18dB NFB gives input sensitivity up around 2.4V-7V RMS! Since typical hi-fi power amp sensitivity is around 1V-1.5V RMS, you won't get very loud.
 

rdf

Member
2004-06-21 8:04 am
big smoke
john65b said:
I opened the chassis and notices one of the three exterior huge CAPACITORS wasn't even connected - marked Mallory 20000 MFD.

What voltage? The 2400 MFD 450 volts Mallorys in my stock are approximately twice the volume of a can of pop so it's hard to envision those as part of a 6l6 HV supply (unless the amp is on castors.)
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
OK, so looks like I may have treaded on Hollowed ground here. Forgive me for being curious. Again, Tube amps are not my FORTE, just find them interesting.

I didn't think I would get so many responses so quickly. I truely understand the dangers of working on these tube amps and appreciate the concern. Bottom line is I don't want to poke around any more than I already have.

The TRANS Tube, well I meant DRIVE Tube.

Anyway, I attached a pic of the amp and set up a website of more pics.

http://suburbansurvey.no-ip.org/DIYTubeAmp.htm

Anyone familiar with the design? Any comments are appreciated.

Good thing is I got it to make music.
 

Attachments

  • front1sm.jpg
    front1sm.jpg
    66.1 KB · Views: 1,856
Delightful project

A guy could have some serious fun with that thing. Why not spend a few evenings drawing the circuit diagram. It would be pretty straight forward with point to point wiring and big parts like that to draw it out. Then we would know just what you are starting with. It is possible that some of the old hands here can peak at those pics of the underside and figure out the basic topology from that.

mike
 

rdf

Member
2004-06-21 8:04 am
big smoke
Don't let stereotypes set false expectations, hallowed grounds and SS vs. tube chips-on-shoulders (pun intended) are rare in this group in my experience. You'll find patience and politeness are rewarded with same, plus an unbelievable bonanza of expertise if you're truly interested and willing to put in the effort.

Voicing an opinion from the low end of that expertise, from what I can see the 20000 MFD is rated 25 volts. No way to tell what the original designer had in mind, possibly DC heaters? It wouldn't suprise me if it's doing no more than sealing a hole after a change in plans. My guess at output iron is Hammond (http://www.hammondmfg.com/1608.htm), you'll need a DVM to guess at the model as the tags are gone. They rate many for use with 6L6 PP so it might not be easy. Ultralinear mode? Hard to tell from the wiring. The power supply trannies are hidden but the PS caps appear to be 120 uF @ 450 VDC. A start at your low power problem would be to measure the DC voltage across the 120 uF caps. It's also impossible to discern how the output transformer secondaries have been strapped. You could confirm that against the diagrams on Hammond's site. Maybe it's set for 4 ohm?

To be honest, if I were in your friend's postion I'ld consider this a kit or a bonanza of good parts (those shiny black resistors look like Mills for example) rather than an amp repair. There are obvious signs of overheating on the perfboard holding the four 100 uF 350 volt caps fed by the rectifiers. If those are inductors under the chassis it's possible they're under-rated for the current required. The wiring can be greatly improved and the original builder made the unfortunate decision to use the steel chassis as a ground plane and signal return. All the hard work and expense are already done and schematics for ultralinear PP 6L6 amps aren't rare. Like mashaffer says: "A guy could have some serious fun with that thing."
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
Thanks for the vote of confidence.

The unconnected Blue Mallory Cap is 20000 MFD and is as you guessed, rated 25v. You may be correct, just may be there to plug the hole. Wouldn't a plastic grommet be better than a somewhat expensive cap?

I took the dimensions off the Hammond website. The Transformer, if truely a Hammond, would be a 1650N or 1650P - as the dimensions jibe.

The 1650N Push-Pull Par. for 2 or 4 tubes and tube type 6L6GC.
The 1650P Push-Pull Par. for 2 tubes and Tube Type 6L6GC.

As far as the history or possibility of schematics for this monster, well I wouldn't hold your breath.

Now, how I got it to work.

I put a 1/2 Amp fuse in the fuse socket as marked (since there was none), and attached speakers. Turned her on and poof, no more fuse. I opened her up expecting to find two fuses, one for the tubes and the main. There was only one. The power cord enters the chassis at a location where the main fuse once was (marked 3 AMP). So I followed the line from the PC and noticed it went right to the 1/2 amp tube fuse socket. So I took a gamble and put in a 3 AMP Slo Blo from Rat Shack, and the rest is history. A ballsy move for not even owning the amp, but I trusted my engineering intuition. Bad thing is the tubes are not protected by a fuse. Looks like the designer had bastardized the amp a bit before dumping it on my friend.

Anyway I tested two different speakers, one 8 ohm and the 4 ohm Maggie. Like I said, sounded very good, warm and tight, but not as loud as I would expect a 40 wpc amp to be. No hum or major distortion when volume knob turned all the way up on the preamp.

As far as measuring the volts accross the two 120 uF power supply caps, what voltage would be correct?

Would worn tubes cause it to be underpowered?

Again, thanks for the help all
 
I think part of your problem is speaker connection. The O/P trafos either have taps or a wiring scheme that allows them to be used with speakers of several impedances. Those power hungry "Maggies" MUST be connected to 4 Ohm taps. Thankfully, "Maggies" have a "flat" impedance curve that mates well with the relatively poor damping factor of a tube amp.
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
> I put a 1/2 Amp fuse in the fuse socket as marked (since there was none)

It probably runs 120 watts idle, more at start-up, so a 1/2A fuse dies fast. 3A is not a bad value, assuming 120V power.

> Bad thing is the tubes are not protected by a fuse.

We rarely did that. Tubes are hard to kill, easy to replace, and (unlike transistors) rarely vaporized by load accidents. We didn't always fuse the Line, but in any DIY project a line-fuse is a must-have.

The driver circuit around the 6SL7 looks OK, as far as I can see. Can't read some values, can't be sure where some leads go, but it isn't wacky. The unit on pins 123 is a voltage amp, cap-coupled to a self-bias Cathodyne on pins 456. The 6L6 are run self-bias, ultralinear. Cathode resistor is a perfectly reasonable 250Ω 7 Watt, bypassed with 220uFd 63V. Grid resistors are 470K, unknown grid-stopper.

I'd say the yellow wire headed "up" is negative feedback. Except it looks like there is a green capacitor on (both) cathodes. If the volt-amp 6SL7 stage cathode is bypassed to ground (if the center lug of each 5-lug strip is grounded) then there isn't any feedback at that point. Yet 6SL7 into 6L6 should be LOUD at normal preamp output levels. Input sensitivity less than 1 volt.

The main power caps are probably that array of four 100uFd 350V at the left. Can't read the values of the 20Watt resistors on that board. 820&#937, 25W? They seem to feed the chokes and then the top-mount caps top-right. That seems to go to the output transformer primary CT, reasonable, though those 20/25W resistors don't seem right. They apparently have been running hot, which means they are dropping more power than the 6L6 cathode resistors. Smells like the military-surplus power transformer put out more voltage than he felt like applying to the 6L6.

The heater leads are not twisted, very bad form. Made worse by using the 6SL7 grid next to the heater-pin as the input pin. While it is silly to feed DC to 6L6, since you have the huge low-volt cap mounted, feeding DC heat to the 6SL7 heaters may be the quickest way to fix any hum (after you get it loud).

Using the chassis as a general ground-dump is bad form. Done right, it can be fine. I'm not convinced it was done right. But let it be for now.

CAREFULLY get the key voltages. I'd like to know 6L6 and 6SL7 Plate and Cathode voltages.

As others say: those Hammonds have tricky output connections, try to verify that it is actually wired right. Tube amps won't suck, 4Ω on the 8Ω tap, but there are dozens of Hammond random tappings that just won't work well for any speaker.
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
I see only two yellow wires - that look like they are coming from the upper Hammond Transformers.

And according to the Hammond website, if yellow is used on the 1650N or 1650P, it would be be wired in an 8 Ohm fashion...

Maybe both the speakers I tried were 4 ohm impedance. Hmm...

I have to say, I did not know a single thing about tube amps three days ago when I brought home this monster. I've read thru a few "Tube for Dummys" articles and am getting more and more interested. Maybe I'll buy the monster from my friend just to continue putzing with it.

It sure beats modding my Sonic Impact T-Amp (hey, it sounds pretty freakin good running on 8 AA batteries and costing me a whole $20!!)

Anyway, anyone want to venture a guess on this amp's value? I want to make sure I don't overpay should I decide to take it off my friends hands (he still thinks its dead).

Thanks.
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
Update

It appears that the Output Transformer (Hammond 1650N or P) is wired for 16 ohm... (i.e. the grn/yel & blk go to speaker out and blk/yel and green are tied. The yellow wire (NFB) goes back "up" to the 6SL7).

I'd say the yellow wire headed "up" is negative feedback. Except it looks like there is a green capacitor on (both) cathodes. If the volt-amp 6SL7 stage cathode is bypassed to ground (if the center lug of each 5-lug strip is grounded) then there isn't any feedback at that point.

Yes, the volt-amp 6SL7 stage cathode is bypassed to ground (the center lug of each 5-lug strip is grounded). Does this mean no NFB is connected? Is this possibly why the Amp doesn't sound LOUDER?

Since I have the Maggies, I rewired the output taps to 4 ohm config (Grn/Yel tied to Green and Blk/yel tied to Blk with Blk and Grn/Yel going to speaker out, and left yellow wire on the 6sl7). I did not hear a noticable improvement - still sounds clear, no hum , but not LOUD for a 40 - 50 wpc amp. And little if any distortion if preamp volume knob at max.

If amp was initially configured to 16 ohm and 4 ohm speakers were connected, this would most definitley stress the amp - could this be why the perfboard looks overheated??

I have not yet checked the key voltages, as I believe the "LOUDNESS" issue with the amp is because of the incorrectly wired Output transformer (to 16 ohm) and NFB possibly being disconnected.

Any more help?? I think I'm getting close to a solution...
 
I wouldn't worry about the discoloration on the perf board. It looks like cheap phenolic pref board was used. I have used it before and it discolors like that when exposed to heat over time. It will even turn brown if left out in the Florida sun for a few months. It's usefulness is not adversly affected, however it will become brittle. It looks like two 10 or 20 watt power resistors are mounted which are expected to get hot.
As far as the lack of loudness, it may be possible that you are just not feeding the amp enough signal. See if you can borrow a preamp, or line stage that will provide more drive voltage. If that is not possible try using the line, or headphone output of a portable CD player. I use an old Sony Discman for amp testing. The headphone jack has plenty of output, which can be regulated with the players volume control. You should have enough signal to drive the amp to slight distortion. Turn the volume control on the amp up full (if it has one) and then slowly advance the volume control on the preamp (or CD player) until distortion just starts to occur. If things are loud then the amp just requires more drive signal that you are currently feeding it. A preamp, line stage, or a re-design of the input circuitry is needed. If distortion occurs, and the amp is still not loud, then there may be a problem with the amp, or your speakers just need some serious juice.
As far as the value of the amp is concerned, there are many variables. For a first guess, look for similar amps on Ebay. Look for amps using the same output tubes 4 X 6L6, and a similar look to it. Remember the Ebay factor, everything on Ebay costs 1.5 to 2 times what it would somewhere else. Also home brew amps tend to sell for less than commercial amps. I have seen home brew amps with similar characteristics sell for $50, a non - working, somewhat ugly, but had UTC transformers amp ( I bought it for the transformers) to $250 for a chrome plated working amp with Hammond transformers. these were flea market sales - not Ebay.
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
Yeah, I thought about that preamp output signal not being sufficient - so I tried the CD player method - as you stated. No go. No more volume than my preamp, and a lot more distortion, so I think it is definitely the amp circuitry somewhere...

Well, now I know about approx value of the amp.

Thanks for the reply.
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
> Does this mean no NFB is connected? Is this possibly why the Amp doesn't sound LOUDER?

NFB reduces gain. If I am reading this right, you are stuck at maximum gain, yet still not happy.

> the perfboard looks overheated??

That cheap under-cured board-stuff is adequate for low-power transistor experiments but inappropriate for 20-Watt resistors and I really would not use it in the same chassis as big tubes.

You have a project, not an amplifier. OK, it works, but 6L6 should blow you away. Something is not right, and it will take some detecting to find out where it is gone wrong.

Value? Much more than $10, nowhere near $1,000. It ain't completely broken but it sure isn't completely happy. The major parts (if good) are worth well over $100, but it needs to find a Good Home to be worth anything at all. A price between $50 and $250 could be fair, depending as much on personal issues as technical and market values. If I was in the mood for a 4x6L6 rip-up project, and I knew the friend needed money more than me, I might pay high; OTOH I'm not in that mood and this isn't my friend, so I might not offer $50 even if he came to my door.

I think it has potential. The power transformer is mil-spec, although it may be a higher voltage than we want. The Hammonds are solid output iron: not virgin-silver audiophile stuff, but mostly fault-free. I hate mechanical hackery, so if the big parts are nicely mounted that is a bonus, even if I have to rip-up all the little parts and wires and start over. But with a full 3 days of tube experience, this will be a Learning Experience.

> configured to 16 ohm and 4 ohm speakers were connected, this would most definitley stress the amp

No. Or not too likely. Tube amps run hot all the time. Some run cooler when working hard. Working into a near-short would not run cool, but on speech/music (especially "not loud") it would not "stress the amp" enough to care about.

> I have not yet checked the key voltages, as I believe the "LOUDNESS" issue with the amp is because of the incorrectly wired Output transformer (to 16 ohm) and NFB possibly being disconnected.

Please survey those dang voltages. Plates and cathodes of 6L6s, all plates of 6SL7. The impedance and feedback issues are probably small. From your descriptions, the driver stage may be woefully misbiased.
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
Good News

I have tracked down the Designer of my sick Amp. He is a retired EE who has built tons of tube amps in the past as hobbies...I will be visiting with him later next week.

I will keep you all posted - I may even get real lucky and get a tube amp education, a schematic of his design, and a fixed - fully functional tube amp out of it!

Seems like a good guy...says he has at least four or five of these amps he built laying around. Maybe we can trade a working computer (I have 11 computers and 3 laptops - computers ARE my FORTE) for a working, high powered tube amp....

I willl refer him to this board, if he doesn't already know of it.
 

john65b

Member
Paid Member
2005-01-09 2:32 am
Chicago
More Stuff...

Well, I finally met with the builder of the amp and he indicated that in its current configuration (ie, Push-Pull, dual 6L6GC, 6SL7, 5R4GYB and the two Hammond 1650N output transformers), is good to go for only about 20 wpc.

Find it hard to believe this amp that weighs about 50 lbs and has all this hardware is only 20 wpc. It's clean, and sounds nice, but doesn't cut it with my maggies.

Anyway, Unless there's a simple mod (different driver or rectifier tubes) that would make this baby sing a bit louder, I am going to return it to its owner...

Any suggestions?. Would those plate & cathode voltages for the tubes matter at this point?

Thanks in advance all....
 
Find it hard to believe this amp that weighs about 50 lbs and has all this hardware is only 20 wpc. It's clean, and sounds nice, but doesn't cut it with my maggies.

Anyway, Unless there's a simple mod (different driver or rectifier tubes) that would make this baby sing a bit louder, I am going to return it to its owner...

Any suggestions?. Would those plate & cathode voltages for the tubes matter at this point?

Thanks in advance all.... [/B]

Hi there.......... With a good LS ....A 20wpc tube amp can blow your socks off......I've used a sim and many MI users remarked it poked near a 50watter. In the 1960's most tube amps were in this power bracket. Most designed for 20wpc with a decent psu will stretch to 30 wpc.

You mentioned in the beginning ...quote; < all tubes had a nice orange glow> WHAT part of the tubes was actually glowing.....the outer or the inner .......OR both ??


richj
 
20W is a bit low for Maggies, but if the builder can demonstrate that the amp is working properly (or can make it so), you might consider keeping it around for the day when you have a more suitable pair of speakers. Tube amps will always be cool and unusual.

BTW, if you think that the pounds per watt tradeoff is poor here, a friend of mine built a 6 watt amp that weighed over 60 pounds.
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
Be nice to know if it worked properly. Some small flaws can make a 20W amp work fine to 2W-3W and then punk out.

> Find it hard to believe this amp that weighs about 50 lbs and has all this hardware is only 20 wpc.

It may be 20W to <20Hz. Which may do you no extra good, if your program materal and speakers only go to 50Hz.

The power transformer appears to be military surplus, which means it is overengineered by commercial standards, and is unlikely to the the exact-right rating for this amp.

Hard to know without studying it in person, but I'd guess if you did a full bore-and-stroke hop-up (sand-state rectifiers, fixed bias, optimized load impedance) there might be over 40 watts on tap. But if 20 watts is not enough, 3dB more is still not enough.

You may just need better speakers. (Ducking while all maggie lovers attack me....)

The biggest hi-fi tube amp I have, a rare Dyna, is 120 Watts, 7.8dB more than 20W. 8dB is just barely "twice as loud", and usually won't change disappointment to bliss. If so, then these Maggies must have taken advantage of cheap sand-state power and 200+W amps, and may not be thrilled on any home tube amp. (I had some commercial 200 Watt {90 lb} tube amps, not really suited for high-end hi-fi. However in efficient speakers they fed crowds of 1,000+ outdoors at party-level. "Good" speakers are worth 10X even 20X of amp watts.)