Liverpool Circuit - Princeton Chassis - 7 pin output tubes

[RJ Guitars]

...Something that you will learn to do after hours and hours of staring at a scope is be able to look at a waveform and know where the harmonics are at....

Not there yet myself. Maybe after we tame this a bit I we can look at some of the high volume waveforms. There is a lot going on in them.

...As far as getting down biasing, it's all a matter of PRACTICE. You need to practice going from paper, building, and measuring your results and comparing them to your calculated results (you did calculate your gain and bandwidth, right? ). L

Not a chance... i built it by the Liverpool schematic except I put in a big enough cathode bias resistor to make me comfortable believing that I wouldn't start by ruining the output tubes. I expected it to have issues but I also expected to find and solve those issues. Right now they look measurable and solvable.

I also reference Merlin's site a lot for quick reminders and such, he's probably my other most often visited site.

I just received a copy of Merlin's book on preamp design. I was immediately humbled at how much I didn't know as I had to slowly and repeatedly read some of the material. I haven't visited his web page but sounds like a good resource.

Simulation is also a powerful tool...

Haven't been down that road either. Might be where I start to draw the line for myself between the hobby and career. I mostly like to solve problems in a way that maybe the common folks can do for themselves. I don't knock the use of advanced tools and skills and I'm not opposed to making use of the data they provide. I am working on myself to make use of the scope more often. Educated guesses and good ears have helped to create some excellent guitar amps but using something as in expensive as a scope (I think mine was $125) will take a bunch of guess work out of solving problems... preaching to myself!

In working through the phase inverter optimization that will be required to bring this amp into line I found some of the easiest to follow explanations and guidelines are from Randal Aiken. Amazing that this stuff is free to make use of. I really appreciate this guys willingness to share his knowledge. http://www.aikenamps.com/LongTailPair.htm

Thanks for all the input on this, I am greatly appreciative of your willingness to work on this with me.

rj

[burkeee]

First thing to check with the headroom issue is the plate voltages in the preamp. If you're running a 250V B+ for the 6AQ5s and using the dropping resistors from the liverpool with a 330V B+, your going to be running the plates really low in the preamp.

[RJ Guitars]

First thing to check with the headroom issue is the plate voltages in the preamp. If you're running a 250V B+ for the 6AQ5s and using the dropping resistors from the liverpool with a 330V B+, your going to be running the plates really low in the preamp.

Doh - you caught me at something that I forgot about until you mentioned it. I used 9.1K - 4.2K - 4.2K in the dropping string. I had used a similar dropping string in my sub-miniature tube Rocket build so this wasn't just a guess. In both cases I wanted to keep the preamp voltages up more than the traditional Trainwreck dropping string would have allowed. https://tubeamparchive.com/viewtopic.php?t=9816

B+1 = 257v
B+2 = 251v
B+3 = 224v
B+4 = 215v
B+5 = 208v

V1b = 118v
V1a = 143v
V2a = 193v
V3a = 176v
V3b = 173v
V4 = 250v
V5 = 250V

You bring up a good point though, before all is said and done I need to scope the the preamp sections and verify what I have there. I went directly for the output because I was intrigued by Zippy's on the spot analysis and wanted to see how well he had sorted this out from my description of the behavior... not bad! I'll be curious if the "major" issue in bringing out some added headroom is the phase inverter?

In the case of the little rocket I fixed most of the problems before I showed anybody the end result. That amp is a much simpler circuit than this one so less fixing was required. I may go back and look at that one again after my education has been advanced a bit by this Liverpool build.

thanks for the input!

rj

PS - Your mention of consideration of low voltages in the preamp brings up something I am curious about and begs the question "do we believe power scaling preserves the tone while uniformly dropping the voltage across an amplifier?" I put this as a PS because maybe this is worthy of an entire thread on it's own ... or put otherwise seems like a self inflicted opportunity for thread hijacking

[burkeee]

Compared to the voltages on the liverpool schem you are running between 30-40V low throughout the PI and preamp. With the Rks from the liverpool and the low voltages, you're running the tubes closer to cutoff than intended.
But then the PI in the liverpool is 269V, which is higher than your B+ even. You could leave the dropping string as is and re-bias each section to suit the voltages you have. That should cure the hard clipping on one side of the waveform from running the bias too close to cutoff.

[RJ Guitars]

Compared to the voltages on the liverpool schem you are running between 30-40V low throughout the PI and preamp. With the Rks from the liverpool and the low voltages, you're running the tubes closer to cutoff than intended.
But then the PI in the liverpool is 269V, which is higher than your B+ even. You could leave the dropping string as is and re-bias each section to suit the voltages you have. That should cure the hard clipping on one side of the waveform from running the bias too close to cutoff.

Thanks for the guidance, I am greatly appreciative.

I added trimmer pots on the PI bias and tail resistors as suggested by the Aiken article and then input my sine wave. For a first dial in I ended up with 520 on the bias resistor and 25K on the tail resistor. It helped a bunch looking at the scope output. At max volume I have a square wave rather than the ugly ragged trace i had before, but still not quite ready for prime time.

I looked at the V1a output - looks perfect... not sure it really is but it's not causing any big issues for now. On v1b as I run the volume up above 10 o'clock I can see one side of the sine wave start to square off, so I'll rebias that tube next then rescope.

Thanks again,

rj

PS - I'll hijack my own thread to some degree but I think it's relevant. If the bias point on the preamp tubes shift with voltage (which it should) then how does power scaling work when you shift the voltage across an entire circuit?

[martin manning]

I'll hijack my own thread to some degree but I think it's relevant. If the bias point on the preamp tubes shift with voltage (which it should) then how does power scaling work when you shift the voltage across an entire circuit?

The bias point moves, but if it was centered to begin with, it stays pretty well centered. The loss of headroom occurs because as B+ is reduced there is less room for +/- signal swing before you reach saturation or cut-off.

[Zippy]

Compared to the voltages on the liverpool schem you are running between 30-40V low throughout the PI and preamp. With the Rks from the liverpool and the low voltages, you're running the tubes closer to cutoff than intended.
But then the PI in the liverpool is 269V, which is higher than your B+ even. You could leave the dropping string as is and re-bias each section to suit the voltages you have. That should cure the hard clipping on one side of the waveform from running the bias too close to cutoff.

Thanks for the guidance, I am greatly appreciative.

I added trimmer pots on the PI bias and tail resistors as suggested by the Aiken article and then input my sine wave. For a first dial in I ended up with 520 on the bias resistor and 25K on the tail resistor. It helped a bunch looking at the scope output. At max volume I have a square wave rather than the ugly ragged trace i had before, but still not quite ready for prime time.

How does it sound? Do you ears correlate to your eyes?

[raiken]

...I found some of the easiest to follow explanations and guidelines are from Randal Aiken. Amazing that this stuff is free to make use of...
rj

Well, I do accept contributions in the form of cash, check, money order, hot Asian mail-order brides, etc.

Actually, all kidding aside, thanks for the nice comments, I'm glad you are getting some use out of the info on the website!

Randall Aiken

[RJ Guitars]

Randall,

thanks for looking in on us. I appreciate your willingness to teach and appreciate your style.

rj

[RJ Guitars]

Tonight I decided to look at all the stages and see what the amp was doing starting at the first gain stage. The scope I use has a built in signal generator so that is what I use for my source. I disconnected the negative feedback before I took any scope shots.

The output of V1a looks like an amplified sine wave and I don't see any distortion.

The output of v1b with the volume set at 9 o'clock maintains the sine wave in it's original form. However, when I start to move the volume up to 12 o'clock I start to see some distortion at the top of the wave form and this continues as i run the volume all the way up to the max. I can guess this isn't good but I beyond that I am not sure how to interpret it. I'll go read some more now and if any of you experienced gurus willing to interpret these results don't mind sharing your wisdom, I'd be pleased to learn something from you.

My apologies for the inconsistent photo quality...

thanks,

rj

[RJ Guitars]

I have worked this amp over several times since the last post... I tried a bunch of different 7 pin output tubes (tons of options available) and finally decided to use the EL95 and have been trying ever since to dial this amp in using them. It had potential for some great tones but was generally over the edge of instability with a thin nasally ice pick sound. I was so confident I could make this work that I had a custom cabinet and faceplate made... it really took me for a long ride but I have tamed the beast:

1) I went back to all the original component values from the Liverpool schematic with only a couple subtle exceptions. It helped but still sounded Harsh. I had tried all kinds of things including a bunch of trimmer pots. I just pulled them all out and put in simple components.

2) I tried an output transformer snubber that kills high frequency oscillations... something featured in Guitar Player magazine that I'd used effectively with poor quality EL34 tubes before. In this amp it killed a lot of tone and didn't fix the problem. I took it back out.

3) Installed lower gain preamp tubes including a 5051 for V1 and a 6072 for V2. Not as crazy but still lots of high frequency hash and ice pick in your ears tone.

4) I then found a few things mentioned inside TAG that transformed this amp into an entirely different animal (in a good way):

https://tubeamparchive.com/viewtopic.php?p=235160
https://tubeamparchive.com/viewtopic.php?p=219333

a. 250pF cap across the phase inverter anodes - really smoothed things out.

b. Installed .033uF PIO Russian caps in the phase inverter... I wasn't sure I could hear any effect but I was inclined to think it cut some of the harshness. I'd heard these were great sounding caps and added a lot of smoothness to an amp.

c) Added a 100pF bypass cap (I used Poly-Styrene caps) on the V1a plate load resistor... The impact of this was huge... I liked this so much I added them to all of the plate load resistors on the other gain stages. If I wanted to keep a little more of the wild child int his amp I would say that only the first stage was essential for this particular amp.

d) Finally I read a 47pf snubber from the anode to the cathode one the preamp tubes. The logic here was that the frequency of noise this kills is far above the audible but messes with the stability of your amp... not a huge effect but maybe audible.

e) I changed the bright caps to 50 and 100 pF Poly Styrene... the 250 and 500 pF from the original schematic were just way over the top. I can effectively use and actually enjoy the bright function now.

The cumulative impact of this created a great sounding amp with sparkling clean tones and a decent solid bottom end and still retained the clean to mean thing with a really rich overdrive tone. I can still sense some instability if I push everything and stand in the wrong position with my strat. I haven't been able to really turn it up and blast away for any long time periods yet but I like it so far.

This was an unusually unstable Wreck build - maybe in part because I broke the component layout from the original but I think it was because these little EL95 tubes were just getting hammered by a huge ugly signal and they couldn't handle it. I need to go back with the scope and look at the waveforms but without knowing otherwise, I think the tone is fantastic. I would say that for the average Wreck this might take too much edge off but I wanted to see how far I could go in taming the beast. The lower volume clean tones are just amazing and unlike anything I have ever heard from a Wreck.

[M Fowler]

Nice rj it looks great.

I have used one of those chassis and wish you would make more for sale

[dorrisant]

Persistence pays off! Greta looking build rj.

Tony

[RJ Guitars]

Persistence pays off! Greta looking build rj.

Tony

Thanks Tony... This one really kicked me around before I sorted it out to the point I did. Now that I have it cleaned up I want to understand what I did so I think I will probably remove each of the added components one at a time and learn a bit more about what they do.