2xKT88 in UL Twin?

[potatofarmer]

The net advantage of DL operation isn't immediately obvious, though after doing a fair amount of reading it does look like Fender did this just to maximize power output.

Don't forget, they also saved the cost of a choke. Maybe they gained a small bit of power, but I'm thinking raising the plate voltage to the 500 ballpark contributed much more to that end.

PaulP, given how easy it is to take these types of measurements, I'd like to ask you to take them.

...would've been posted before you did if the cat didn't step on the F#*#(* plug strip!

My measurements suggest that it's inappropriate to refer to the Super Twin Reverb as an "ultralinear" amp. It would be interesting to see some similar measurements off of an "UL" Twin Reverb. It may be that we need to put an end to a bad habit in referring to these amps as "ultralinear."

My measurements agree with that assessment.

The plate-to-plate impedance is a little odd; I wouldn't expect Fender to deviate much from their standard 2k (though DL might require it?). I had a pet theory that this OT was actually designed for a pair of KT88s with 8, 16 ohm taps, and the KT88 datasheet lists 4.5k as one of their sample Ra-a conditions for UL connected amps, but then the screen taps don't make sense, so it probably was spec'ed by Fender for the tubes and output taps found in this amp.

[potatofarmer]

I had to re-read what I posted on the old thread and bob p's OPT measurement. Two things that you should consider (if you have not done so already):

- the Fender OPT has 12.5% tap to maximize power
- the recommended UL tap for KT88 is 40%

So it is not quite a straight-forward substitution. The only thing we are missing is the primary impedance, since bob p's OPT is for a sextet not quartet, with that information, we can draw some load lines to see how well the KT88 will perform in the amp. If someone has the spec on the OPT, please post it.

Below is the UL distortion and power output characteristics for the 6L6:
[ATTACH=CONFIG]26925[/ATTACH]

GAH, I keep losing my posts.

I think that yellow box is misplaced; I'm measuring the screen taps at 12.5% of the winding, not 12.5% of the plate impedance.

Here is a similar plot for the KT88:


So, at 12.5% of the winding we gain ~2W (whee!) but the output impedance and distortion have already dropped significantly. Pretty far from the optimum 40%, but if I'm reading these right, KT88s would be closer to true ultralinear operation than 6L6GCs at these screen taps. I'd have 8 and 16 ohm taps instead of 4 and 8; no big. Gotta replace the speakers anyway, unless mold makes a Utah cone sound better (it doesn't.)

[jazbo8]

Good catch, sorry about the 6L6 graph, I need to fix it... I don't think Fender ever intended the amp for true ultra-linear operation anyway, so you don't need to either.

[bob p]

With modern line voltage my B+ is significantly higher than +500 VDC. That's pretty hard on 6L6. That has given me pause to think about redesigning the power amp around more robust tubes. The 6550 and KT77 both came to mind, though a full set would require some power supply mods to deal with the increased heater current requirements.

I don't know about the physical fitment of the larger sized bottles in your Twin Reverb, but 6550 will "fit" in my Super Twin. That is to say I can plug them into the sockets, but in doing that I would be violating both good engineering practice and the specs for on-center distance with passive cooling. Mick's point about inverted operation should not be overlooked. IMO these amps already run too hot with 6L6; my amp came with baked socket-mounted CC resistors in the power amp that were causing thermal noise. If there were ever a group of amps that would benefit from improved ventilation it would be the multiple-pair inverted Fenders.

Regarding that plot, I think this is a better representation of what you were thinking:

[Alan0354]

Call me ignorant, you want ultra linear and you know you are not going to run the twin into distortion. What difference if you use an amp with SS power amp? SS power amp is literally a discrete opamp with much higher open loop gain that correct most if not close to all the distortion. You cannot even talk linearity of tube power amp with SS. No matter how you cut it, the open loop gain of the tube power amp in Fenders are so low you literally can open the loop and the volume might just a few times only. YOu can't talk linearity with this low open loop gain!!!

More importantly, what make you think the amp sounds good if it is totally linear?

[Enzo]

I cannot speak for everyone participating, but make sure not to confuse Ultralinear as a measure of performance, and Ultralinear as a description of circuit topology. Speaking for myself, when I describe an old Fender amp as "ultralinear" I am talking about it having a tapped OT primary, I am not making any claim about its fidelity. The uses of ultralinear circuits in hifi are not mysterious, but guitar amps are not hifi.

[potatofarmer]

With modern line voltage my B+ is significantly higher than +500 VDC. That's pretty hard on 6L6. That has given me pause to think about redesigning the power amp around more robust tubes. The 6550 and KT77 both came to mind, though a full set would require some power supply mods to deal with the increased heater current requirements.

I don't know about the physical fitment of the larger sized bottles in your Twin Reverb, but 6550 will "fit" in my Super Twin. That is to say I can plug them into the sockets, but in doing that I would be violating both good engineering practice and the specs for on-center distance with passive cooling. Mick's point about inverted operation should not be overlooked. IMO these amps already run too hot with 6L6; my amp came with baked socket-mounted CC resistors in the power amp that were causing thermal noise. If there were ever a group of amps that would benefit from improved ventilation it would be the multiple-pair inverted Fenders.

I know, right?! 6550s are probably a better size bottle for this amp than KT88s, what with the PT pretty close to the sockets, but that oestex guy warns about running them in ultralinear; apparently the screens aren't as robust as the KT88. Then again, with current production tubes you could buy a KT88 and get a 6550 anyway...

So the apparent options I see are:

• Use large screen stoppers on 6L6GCs and just ignore the poor conditions.
• Convert to a different tube type and stay with ultralinear mode.
• Convert to a different tube type and convert to pentode mode.

I think 6550s would be a good candidate for the third option, and we have a convenient node for 1/2 B+ right where the PT CT connects at the mid point of the stacked reservoir caps! It was speculated in a previous thread about these amps that you can actually draw current from that node without disrupting the voltage balance across the caps and indeed Fender did just that a few years later for a 25W "low power" option on a Twin. Just add a choke (or a dropping resistor) and a new filter cap, and you're good to go.

Plotting the load lines for B+ = 510V, Raa=2.26k (or 4k5) at Vg2=255V should show whether 6L6GC or 6550 would be a better choice. Of course, going to a quad of larger tubes will require additional heater current, in addition with the poor airspace between each tube. No idea what do do with three to a side.

I'm personally still leaning towards a pair of KT88s. That guy on the steel guitar forum said that they fit, and the outer sockets should be far enough apart. Of course, if one plate seam winds up perpendicular to the back of the amp, the other seam will be pointing right at the PT which I'm sure will bounce the heat right back. And heating the PT isn't such a great idea either. Perhaps a thin Al sheet to reflect some of that heat away is in order.

Again looking at the old Genalex KT88 datasheet (guess I should finally link to it; this one: http://www.r-type.org/pdfs/kt88.pdf ) to get 100W out of a pair in ultralinear the plate voltage would have to be around 560V. Sooo the switch to KT88 will possibly reduce the power output, on paper at least.

Regardless of any tube swapping, adding a fan is certainly a good idea.

[potatofarmer]

What difference if you use an amp with SS power amp?

Well, I already have one of those. An old Kasino Concert, 100W, sounds great, even breaks up nice. Converting this thing to a SS amp would be quite a bit of work though. And I certainly plan on trying to push this thing into distortion.

[jazbo8]

Call me ignorant, you want ultra linear and you know you are not going to run the twin into distortion. What difference if you use an amp with SS power amp? SS power amp is literally a discrete opamp with much higher open loop gain that correct most if not close to all the distortion. You cannot even talk linearity of tube power amp with SS. No matter how you cut it, the open loop gain of the tube power amp in Fenders are so low you literally can open the loop and the volume might just a few times only. YOu can't talk linearity with this low open loop gain!!!

More importantly, what make you think the amp sounds good if it is totally linear?

I cannot speak for everyone participating, but make sure not to confuse Ultralinear as a measure of performance, and Ultralinear as a description of circuit topology. Speaking for myself, when I describe an old Fender amp as "ultralinear" I am talking about it having a tapped OT primary, I am not making any claim about its fidelity. The uses of ultralinear circuits in hifi are not mysterious, but guitar amps are not hifi.

I'm not sure if Alan0354 is using "ultra linear" as an adjective, it he is, then Enzo's comment is spot on. OTOH, just because an ultra-linear output stage is used, it does not mean it has to be linear and has low distortion, you can still over-drive it and get power tube distortion.

[Alan0354]

I'm not sure if Alan0354 is using "ultra linear" as an adjective, it he is, then Enzo's comment is spot on. OTOH, just because an ultra-linear output stage is used, it does not mean it has to be linear and has low distortion, you can still over-drive it and get power tube distortion.

Well, by definition, linear means the transfer ratio is constant.............Straight line response. This is a definition pretty much accepted by electronics describing linear and non linear. Linear pretty much defined as

y=kx where k is a constant. That's linear. So ultra linear can be interpret as very linear!!!

Back to my comment, I read something about tapping OT to drive Screen grid to get more linearity out of the power amp, that's what I comment about. Why do people want a very linear guitar amp? I don't think it sound that good if the amp is totally linear. It's the tube distortion ( not over drive type) that gives the character of the tube amp. Also, a Twin is so high power it's not practical to use it in clipping distortion by cranking it all the way.

These ultra linear OT tapping is much more common in hifi amp than guitar.

[jazbo8]

So ultra linear can be interpret as very linear!!!

That's the point Enzo tried to make, the world "Ultra-Linear" is just a name that Hafler & Keroes came up with for marketing purpose (pretty clever!) - it is not an actual electrical term as you seemed to take it.

[Alan0354]

That's the point Enzo tried to make, the world "Ultra-Linear" is just a name that Hafler & Keroes came up with for marketing purpose (pretty clever!) - it is not an actual electrical term as you seemed to take it.

But it is supposed to be more linear than fixed screen voltage. It still seems to go the opposite way of a good tube amp ( distortion).

I never listen to this particular Twin, so I cannot comment on the sound. But most of the famous/popular amps don't do this. From my understanding, this is not a popular amp by any stretch. What is so special about this amp? Does it sounds really nice compare to the much more popular regular Twin? If it does not sound as good, maybe it's time to just put it at fix screen and make it a regular Twin that sound mighty good.

[jazbo8]

But it is supposed to be more linear than fixed screen voltage. It still seems to go the opposite way of a good tube amp ( distortion).

I never listen to this particular Twin, so I cannot comment on the sound. But most of the famous/popular amps don't do this. From my understanding, this is not a popular amp by any stretch. What is so special about this amp? Does it sounds really nice compare to the much more popular regular Twin? If it does not sound as good, maybe it's time to just put it at fix screen and make it a regular Twin that sound mighty good.

I think/speculate when Fender made the Super Twin, the ultra-linear amps were the latest fad in hi-fi land, so why not hop on the bandwagon... The fact that most guitar players prefer pentode power tube distortion (over that of the triode), might be another reason why UL never caught on for guitar amps, like you said, the whole idea of lowering distortion in guitar amps just seems strange... Besides, in the case of the Super Twin, it is not what most people would consider true UL anyway, so it appears to be a historical oddity.

[Alan0354]

I think/speculate when Fender made the Super Twin, the ultra-linear amps were the latest fad in hi-fi land, so why not hop on the bandwagon... The fact that most guitar players prefer pentode power tube distortion (over that of the triode), might be another reason why UL never caught on for guitar amps, like you said, the whole idea of lowering distortion in guitar amps just seems strange... Besides, in the case of the Super Twin, it is not what most people would consider true UL anyway, so it appears to be a historical oddity.

Ultimately it's the sound.....which I have never heard before. If it is good, use it. If it is not as good, I'll change to fixed screen and be done with it!!!

[potatofarmer]

Ultimately it's the sound.....which I have never heard before. If it is good, use it. If it is not as good, I'll change to fixed screen and be done with it!!!

Yep! One benefit of the distributed load operation is that it's also a form of local NFB, so you can get rid of the global NFB loop and its associated problems with phase shift. Fender added a "speed up" capacitor in parallel with the feedback resistor, but I imagine that has as much to do with their terrible lead dress as it does with applying more NFB than typical for guitar amps.

And, of course, you can't mention UL/DL in guitar amps without bringing up Dr Z, who's had some success in making UL guitar amps. Though now I really wonder where he specs his screen taps...