The "$250" Amp? - 6LU8 Compactron SE, UL

Anybody see this? I think it's in Norwegian or something. But it looks like 4 6LU8's used in push-pull fashion daisy-chained to make 2x12.5 W and a nice looking square wave. I realize we're aiming for 2 tubes single-ended, but I just wondered if this helped at all.

http://www.veiset.net/6LU8/index.html

6lu8_amp_ii_r.gif


Note: Also, while poking around the web, I saw someone comment that he thought the power part of the 6LU8 was more like the 6W6, and someone else commented that the 6W6 was a very under-rated tube for stereo amps. Just thought I'd toss that in there. You guys can feel free to tear it apart if that's not true.
 
You mention one way of triode rigging a pentode. It is quite different form the g2-plate version. See the RCA amps running a type 46. The intended triode version for this amp is the g2-plate version, and the resultant mu is listed as mu g1-g2 in the data sheet( it is 6.5 ).

The g2-g1 strapping can be examined in the sheet for type 3D21WB from Tung-Sol, note the radically different plate curves between the two arrangements.

What Douglas is saying here is that a triode is composed of three elements, we will call them the Grid, the Cathode, and the Plate (anode). This is the minimum number of logical elements that a signal tube, like in an amplifier may have to do it's function. A pentode has two extra grids (G2 and G3), for a total of five elements. A pentode can be made in to a triode by tying those extra grids, G2 and G3, to one of the basic elements G1, Cathode, or Plate.

Douglas notes that I only mentioned the most common way of making a pentode mimic a triode: By tying G2 to G1, and G3 to the Cathode (to form a suppressor grid, something worthy of another post in itself). He notes that in some cases you can tie G2 to the Plate.

What you should take away from this discussion is that most pentodes can be made to act as triodes by connecting their second grid (G2) to the plate via a low resistance resistor (100 ohm or so), and their third grid (G3) to the Cathode directly. Sometimes the G3 to Cathode connection is made internally in the tube itself, and you do not have to worry about it at all. However, there are alternate arrangements where G2 and G3 get connected to different places to make a triode. It is best to consult a particular tube's data sheet, and a particular implementation's schematic in all cases.
 
That helps a bunch, Ed, thank you very much.

Now, with regard to ultralinear mode...

I have been perusing old rec.audio.tubes postings for more information on the 6LU8, and a fellow named John Stewart was apparently quite a fan of them in 2000. He even designed and built an amp that was featured in Glass Audio, if you remember that magazine. He says he was able to put it in UL mode quite easily. I quote:

The Hammond 125 Series PP Output Transformers does work
"A OK" in a SE circuit. I've done that not once but many times
beginning more than 40 years ago.

If you like you can even use the center tap to form an ultralinear
mode connexion. Done that, too!!

If you want to take the time you can even read about how to do it
complete with comprehensive test results in two of my published
articles on SE amplifiers as follows-

AN AFFORDABLE SE TRIODE AMP-
Uses a 6EA7/6EM7 Double Triode into a 125E
Glass Audio Magazine Issue 4, 2000

MORE POWER FOR THE AFFORDABLE SE AMP-
Uses a 6LU8 in Triode or UltraLinear Connexion into a 125E
Glass Audio Projects Book

See www.audioXpress.com for these publications.

There are several other unpublished amps, both SE & PP
using the Hammond 125 Series.

Again, I may be throwing out terms here I don't understand. Just trying to help.
 
Anybody see this? I think it's in Norwegian or something.

Way cool Wiggy! That amp actually uses 2 6LU8 per channel to make a Push Pull amp. You can see on the left of the schematic there are two tubes with three elements each. This would be the triode section of 2 6LU8s. Then, you see that there are 2 tubes with 5 sections each on the right hand side, sort of on top of one another. These are coupled to a transformer. These would be the output tubes in a push-pull arrangement.

Each 6LU8 contains a single small triode, and a larger pentode. The schematic has 2 triodes and 2 pentodes. Even tho each triode and each pentode is represented as a single tube, one triode and one pentode are in each physical bottle, so you only need 2 tubes per channel. (As opposed to one for a SE amp.)

Hope this makes sense.
 
Hope this makes sense.

Actually, it kind of does. I recall how you explained it to me that day we took photos of your amp as you were building it some months ago. With two tubes per channel in a PP arrangement, is the first tube feeding the second, or are they in parallel?
 
Link to PDF data sheet for the 6LU8, in case that's of any use. I know we've got several now

When you look at a data sheet for a single 6LU8, you can see it has a three element triode structure, and a five element pentode structure in it. When you see a schematic that calls for a 6LU8, look to see how it defines the use of the triode, and pentode elements of the tube. Tho they may appear to be different tubes on the schematic (in different places, they each are drawn as "their own tube" in a circle, etc.) they are discreet structures in a signle physical glass envelope.

Man, I'm a dork.
 
When you look at a data sheet for a single 6LU8, you can see it has a three element triode structure, and a five element pentode structure in it. When you see a schematic that calls for a 6LU8, look to see how it defines the use of the triode, and pentode elements of the tube. Tho they may appear to be different tubes on the schematic (in different places, they each are drawn as "their own tube" in a circle, etc.) they are discreet structures in a signle physical glass envelope.

Right, I got that from Kegger's drawing. I realize that the schematic is the schematic, not the actual physical tube, so each tube's separate parts are listed twice. I actually do the same thing when I am doing logical architecture drawings of network components for the data centers I design.

I'm a dork too. Just a computer dork. Same family, different genus.
 
Actually, it kind of does. I recall how you explained it to me that day we took photos of your amp as you were building it some months ago. With two tubes per channel in a PP arrangement, is the first tube feeding the second, or are they in parallel?

Power tubes in a PP arrangement (generally speaking for the moment) each boost half of the audio signal, which is then combined in the transformer. The output tube's job is to take what it's given, and make it more powerful.

The small signal tubes (the triodes) in a PP arrangement are there to split the signal up and hand it to the power tubes. When you see that second triode structure in that Norwegian schematic, look at how there are lines that go from it to each power tube. It is splitting the signal in half (by phase, to be technical) and handing half to each power tube (who then amplify it, and then the output transformer recombines it because it's a PP transformer built for that specific purpose).

In the SE arrangement that we will use, there will be one power tube coupled to the output transformer. There will be a single triode in front of it, for some voltage gain. We do not need to split the signal, so there is no phase splitter. The small triode and large pentode will be in the same bottle.

Again, I hope this makes sense.
 
In the SE arrangement that we will use, there will be one power tube coupled to the output transformer. There will be a single triode in front of it, for some voltage gain. We do not need to split the signal, so there is no phase splitter. The small triode and large pentode will be in the same bottle.

Again, I hope this makes sense.

Yes, it does. My brain is pulsing and glowing like a tube, but some of this is sinking in.
 
sounds interesting, what about those outside the US and run on 240v, can we play too?

Someone can correct me if I'm wrong, but I would imagine that although you'd choose a different power transformer, everything else would be the same. However, don't take my word as gospel.
 
So it sounds like everyone had decided on a design and pretty much how it needs to be wired up? So will we be getting a parts list soon? something that will tell us what kind of iron we need to find? Sources and stuff?
 
hey-Hey!!!,
I would like to introduce a few options to the power TX, and its use and why a small upgrade might be useful.

First, there are three power levels of the 275-0, 0-275 Antek power torroid. From smallest to biggest the difference is $10, with the big one capable of about an amp of current from the HV coil.

Second is tied to the current capacity options, and that is using a pair of tube diodes on a 275-0-275 HV coil or putting a half SS diode/half tube diode full bridge on a paralleled set of HV coils.

If you put the coils in series, we'll only need a single double diode, perhaps a 5AR4 to do the job. A pair of single diode tubes from the damper diode family( such as 6DE4, 6AU4, 6AX4, 6CJ3, 6DN4 ) are quite inexpensive, and in this service are operated very conservatively.

Running damper diodes will take up the supplied 4A 6.3 v winding on the suggested Antek power TX, so a 3A 6.3V tX would have to be added for the two power tubes each needing 1.5A. That could be a Xicon, supplied by Mouser for $7 each. Two damper diodes, and the 3A filament TX are approximately equal to the price of a single 5AR4...:)
cheers,
Douglas

for this one, the triode strapping method is to tie the g2/screen grid to the plate. That'll deliver a triode with mu of 6.5 and plate resistance of ~700R.
 
sounds interesting, what about those outside the US and run on 240v, can we play too?


IIRC the data sheet correctly, it simply requires you to connect the two 120V primary windings in series to operate on 240V. In the states, we just put them in parallel.
cheers,
Douglas
 
hey-Hey!!!,
I would like to introduce a few options to the power TX, and its use and why a small upgrade might be useful.

First, there are three power levels of the 275-0, 0-275 Antek power torroid. From smallest to biggest the difference is $10, with the big one capable of about an amp of current from the HV coil.

Second is tied to the current capacity options, and that is using a pair of tube diodes on a 275-0-275 HV coil or putting a half SS diode/half tube diode full bridge on a paralleled set of HV coils.

If you put the coils in series, we'll only need a single double diode, perhaps a 5AR4 to do the job. A pair of single diode tubes from the damper diode family( such as 6DE4, 6AU4, 6AX4, 6CJ3, 6DN4 ) are quite inexpensive, and in this service are operated very conservatively.

Running damper diodes will take up the supplied 4A 6.3 v winding on the suggested Antek power TX, so a 3A 6.3V tX would have to be added for the two power tubes each needing 1.5A. That could be a Xicon, supplied by Mouser for $7 each. Two damper diodes, and the 3A filament TX are approximately equal to the price of a single 5AR4...:)
cheers,
Douglas

for this one, the triode strapping method is to tie the g2/screen grid to the plate. That'll deliver a triode with mu of 6.5 and plate resistance of ~700R.

SQUIDWARD! He's gone all fuzzy again! Can someone adjust his horizontal hold?
 
Back
Top Bottom