How did you reach stability..?

[ODwan]

Hello everybody!
I'm currently building/tweaking an Express and have encountered the stability-problem. After having read RossH's entry about his conversation with KF I was quite upset! That is exactly my approach: building the amp with virtually no bandwidth-limiting features and if it's oscillating, bring it down just a tiny bit below that point. Has someone here taken the same road before? My question would be: how did you tame the beast without loosing it's wildness?
My checklist would be:
1) Lead dress (moving around wires)
2) Swapping tubes (mainly preamp)
3) Shielding cables (beware of that parasitic capitances!!!)
4) NFB adjustment via the serial NFB-resistor (maybe use a pot for tweaking)
5) Grid stop resistor on the second stage
6) Adjusting the RC-combination between 2nd and 3rd stage

I hope someone can add some wisdom...

Timo

[RossH]

On my wreck I ended up:

1) having to put shielded wire on the first stage
2) having to put shielded wire between the volume and the 2nd stage
3) Changing my second preamp tube (I was using a basic Sovtek and went to a darker sounding National 12AX7.)
4) Putting a 68k grid stop resistor on the first tube

Those four things took care of my oscillation.

However, I built mine in a JTM45 chassis for a short box head. I don't know what impact that may have had on things.

Other mods were:

1) Pot-to-ground MV before the PI
2) Paralleled first preamp tube
3) .0047 cap between 2nd and 3rd stage (more bass)

Ross

[Dai H.]

the Komet pics showed use of shielded wire, so personally I'm not so sure that it would necessarily be bad for the sound.

[Jackie Treehorn]

There was a post on the old site by someone who had been in a wreck who had discovered a 33k grid resistor at the end of the shielded wire to the first gain stage. I used one on my build. I also added a small value grid resistor around 1.5k, I think, on the second stage since I was getting some splatty blocking distortion. I've had no grid resistor to 100k on the second stage and didn't find the large value to audibly cut the highs, but it did smooth the attack character of the notes a bit.

The problem I can see with not using any kind of components to cut some high frequencies is that you end up with a bad sounding amp in my opinion! Really bright with a lot of hiss and "crispiness." An amp with that much tinnitus inducing highs is not that useful to me! I tried bypassing 2nd and 3rd stage plate loads to cut highs and that definitely makes the amp sound dead. To my ear, a small cap across the phase inverter plates cuts enough highs without changing the character of the amp. The Komet has a cut control across the plates, so I don't think Ken would have a problem with that approach. The pictures I've seen of Alessandro's also use a similar cap.

[Omar]

Hey Jackie,
First of all, thanks for the tweaking tips you posted on The Gear Page for your Trainwreck clone. The cap across the phase inverter was a great tip and tamed the highs but kept the character of the amp.

I also found that changing to Dale/Vishay metal film resistors also improved the tone of the amp. I originally built my twreck with carbon comps but it was noisey and lacked something in the high end. I changed to metal films to address the noise but was surprised that it improved the tone.

Omar

[seulb-rd]

[quote="Omar"]Hey Jackie,
First of all, thanks for the tweaking tips you posted on The Gear Page for your Trainwreck clone.

can i get directions or a link to "The gear Page"? i'd like to see what type of things can be to tame the beast.

tia

[Omar]

I saved the Jackie's posting from The Gear Page. I hope Jackie doesn't mind me re-posting it here.

The problem I am having with my Trainwreck clone is that at very high volume (from about 2 o'clock and higher), when you hit a chord with a lot of low end and mute the strings quickly, the normal hiss and noise is gone for about half a second and slowly builds back up. I saw a couple of people mention this when talking about Trainwrecks and Trainwreck clones and asked if this was normal. Both Mook and Hogy said that Trainwrecks should not do this.

With that disclaimer, I'll tell you my thoughts. In order to get the clean sound with the guitar volume rolled off as loud as the overdriven tone, you've got to have significant compression. I think the sound you're hearing with the hiss means you're on the right track. I know my build exhibited this kind of behavior at times, and I think what minimized that behavior (essentially cutting back the hiss and keeping the compression) was the use of grid resistors (which aren't used consistently on the blueguitar schematics), which limit the high frequency response and tend to stabilize the amp, and a 50 pf capacitor across the plates of the phase inverter to cut back the excessive high end and also aids in cutting oscillation (which also is not on the schematics). In terms of the grid resistors, I used a 33k at the input of the first stage, something between 600 and 800 ohms for the second stage (mainly to eliminate some splattiness I was getting), and 1.5k on the power tubes (I had 5.6k for awhile, but the amp is more punchy with the 1.5k's).

I hear the asymetrical clipping of the third stage as compression more so than "distortion." The resistor just prior to that stage, controls how hard the third stage gets hit. A 150k, what I like, hits it hard and gives a lot of compression, probably what is causing the pumping you're hearing after playing a loud chord. A lower value, 56k for example, will compress less, but to my ear with a strat, doesn't get the clean sound the same volume as the overdriven tone. I think it's a good idea to choose the resistor value to give the compression you want with the guitar you want to use, then choose the coupling cap value before the thrid stage to be the lowest value before the bass drops out.

I hope that's helpful!

[DaveH]

the Komet pics showed use of shielded wire, so personally I'm not so sure that it would necessarily be bad for the sound.

Why would shielded wire sound bad? You have 20' of it in front of the amp, so there's no way that an additional 6" is going to adversely affect anything. I use the small diameter George L cable in those sections...it's high quality cable with low capacitance, and it's small enough that it's easy to work with inside an amp. If you use cheap guitar cables, they are going to mess things up much worse than this little bit of extra shielded cable will.

Dave

[tubedogsmith]

the Komet pics showed use of shielded wire, so personally I'm not so sure that it would necessarily be bad for the sound.

Why would shielded wire sound bad? You have 20' of it in front of the amp, so there's no way that an additional 6" is going to adversely affect anything. I use the small diameter George L cable in those sections...it's high quality cable with low capacitance, and it's small enough that it's easy to work with inside an amp. If you use cheap guitar cables, they are going to mess things up much worse than this little bit of extra shielded cable will.

Dave

Try it both ways and whether or not you can hear any difference is the answer for you.

[Jackie Treehorn]

In trying to obtain stability, it's probably important to identify which parts of the amp should be relatively stable and which parts should be close to the edge so that the stable parts remain so and the crazier sections are unhindered. I don't think the first stage is ever driven into distortion by the guitar pickup, hence the grid resistor on that stage basically should block out hiss and radio interference, the non-musical elements, from entering into the stages which generate the musical distortions and harmonics. I think the second and third stages need the extended high end in order to really be on the verge of oscillation. I suspect that there's something in the layout of these stages which I'm not fully grasping. If you look at the layout in the files section, those two stages don't strike me as being laid out conventionally and for the shortest wire lengths. The layout of these stages and wires to the tonestack also seem to have a big effect on the ease of controlled feedback.

Sticking the capacitance across the plates of the phase inverter uses phase cancellation to cut the high end content, however I don't think it would have any effect on the stability or response of the prior stages. It would definitely change the content to the output tubes. I did some experimentation with a couple pieces of wire to see if they could be laid out in such a way as to generate 50 pf or so of capacitance. I think it would be rather hard to get the kind of roll off just with layout.

If you can use an oscilloscope to determine the source of the instability, and use the smallest possible value component to stabilize the stage, you're still on the verge of instability. In that same thread Omar quoted, Hogy had stated you should be able to turn up all the controls on a wreck and not have any artifacts. I upped the value of my treble cut cap on my PI plates to 80 pf a couple days ago because the amp had still been a touch too bright As a side effect, I can dime the volume now without the stray noises it used to make at that setting. I think I'd get some weirdness with the treble and presence dimed, though.

Anyway, in regards to the hiss in Omar's amp, I don't recall anyone on that thread who owns or is familiar with wrecks giving any suggestions for solutions. Unfortunately, I think there are some parties who have the capacity to help, but don't feel it's in their best interests to have hobbyists building great sounding wrecks for a few hundred dollars in parts. I suspect a portion of information volunteered or questions left unanswered is to discourage or even divert. Geetarpicker posted that his wreck showed the same characteristics you described and his wreck is the best sounding one I've heard from the clips! Essentially I wouldn't take everything on the internet, aside from chassis pics, as absolute truth.

I like those dale resistors, too, btw.

[seulb-rd]

thanx, omar!

[Dai H.]

the Komet pics showed use of shielded wire, so personally I'm not so sure that it would necessarily be bad for the sound.

Why would shielded wire sound bad? You have 20' of it in front of the amp, so there's no way that an additional 6" is going to adversely affect anything. I use the small diameter George L cable in those sections...it's high quality cable with low capacitance, and it's small enough that it's easy to work with inside an amp. If you use cheap guitar cables, they are going to mess things up much worse than this little bit of extra shielded cable will.

Dave

I'm saying that maybe the idea that you shouldn't use it should be re-examined being that it is evidently used in amps KF is involved with.

[tubedogsmith]

Anyway, in regards to the hiss in Omar's amp, I don't recall anyone on that thread who owns or is familiar with wrecks giving any suggestions for solutions. Unfortunately, I think there are some parties who have the capacity to help, but don't feel it's in their best interests to have hobbyists building great sounding wrecks for a few hundred dollars in parts. I suspect a portion of information volunteered or questions left unanswered is to discourage or even divert.

Egggsactly!!