Framus Cobra -20dB mod

[Moodknyller]

Cobra_V5-01.jpg

Hi guys, first post ever in any forum here. And I turned 31 today. Better late than never I guess.

So the problem I'm having is the Cobra's loop being 0dB, and I like the channel volume up high,
which makes my dd-500 in the loop fart out. And using preamp bass past 9 o'clock isn't really helping either.

Any advice on modding the loop instrument level?

I've already done some mods to the amp:
- Switch for parallel/off/serial fx loop
- Higher gain on crunch channel
- Notch switch functions replaced with Cobra/Dragon..ish voicing
and I'm about to put a fizz cap in it, so it's not like I'm doing something completely new to me.

Thanks in advance.

[ajaxlepinski]

Welcome to the fold brother! This is the best guitar/amp forum on the interwebznet!!!

Can you add a volume pot to the fx loop send?
The easiest thing to do is to plug in a Carl's Custom volume box in, between the fx send and your effect....
https://carlscustomguitars.com/The-Orig ... me-Box.htm

[clipless bumper]

I had post already to go - but got distracted (at work) - and must not have hit 'submit'......

anyway - there are any number of pedal makers that sell pedals just to switch from line to instrument level.

Ebtech also makes units - here is their web page.

[Moodknyller]

Yeah well, the thing is.. I did consider the ebtech and the g lab equivalent, but I can't justify paying 100+€ for a line level shifter
knowing I can do a small mod in the amp for a few bucks and it does the same thing. Also I like to tinker and learn new stuff.

I'm an electrician by profession, and I've seen how we are being fooled by fancy looking pc-boards costing crazy amounts of money,
like a room temperature sensor which consisted of A SINGLE resistor and two screw terminals soldered to 9x9 cm pc-board inside a 10x10 cm
cheap plastic chassis costing around 50€. I understand they need to make a living, but some things are taken too far.

And yes, I could add a volume pot to fx loop send, but my concern is will there be enough level boost on the return?

[ajaxlepinski]

The DD-500 volume level should be able to compensate for using a volume pot to reduce the fx send signal.

The Carl's Custom Volume box is only US$17.00 + shipping. You could build one for less so, why not give it a try?
Test it externally... if it works, you can install it in the amp.

[Moodknyller]

Okay, I'll try that.
And if I want a return level adjustment should I put something like a
100K or 250K volume pot where R46 is?

[Moodknyller]

So I threw in some B500K pots I had lying around, one between C36 and fx send,
the other one replaced R46.

The return level pot doesn't do much, which makes sense to me now, being that the signal doesn't
go through the pot. It can be made more effective if need be, just needs a bit more fiddling.

The send level pot has a much greater effect, which was expected.
But somehow it also acts as a low pass filter. A really noticeable one too, which is not what I'm hoping for.
Am I missing something here?

[Zozobra]

How have you wired it up? A 500k pot is going to be too big and at lower settings will shunt a lot of treble to ground, much in the same way your gain pot does and why bright caps are used.

This is the send from ye typical engl circuit:


Try that. Also check this for a good series loop on page 219:
https://books.google.co.uk/books/about/ ... &q&f=false

If you just want to read on the send thats covered on pg 215.

P.S. buy the book. It's a must have.

[Moodknyller]

I thouhgt about 500k pot being too big, so I took it out and tried various resistors instead, like 47k in signal path and 4.7k to ground between C36 and fx send.
The affect is always the same, very noticeable loss of highs, bite and attack. Higher master volume doesn't help.

Maybe there is a way to make the amp sound good at lower channel volume?
Right now anything below 3 o'clock is rreeaally fizzy and dull. Dialing gain back to 10:00 does take the fizz away but it's not enough brootz for me.

[Zozobra]

Sloclone has lots of mods for the slo and this is pretty much the same thing so look there for inspiration about the fizz. They'll have some thoughts on the loop too.

First things I'd try are reduce the cold clipper from 39k to anywhere between 4k and 22k. People tend to like 10-12k here. That smooths the high end hash from the cold clipper. You can also try increasing the 4th stage grip stopper from 220k to 470k or 680k to kill some fizz.

[Moodknyller]

So apparently it took me almost a year to solve this issue.

I tried swapping, adding and removing a lot of parts. Some were an improvement, most were not.
I found the smaller cold clipper to add fizz, but I kinda liked the fluidity it gave me so I have it on a switch to go between 12k/17k IIRC.
Changed plate resistors, added plate bypass caps, removed cathode bypass caps, increased nfb etc. and all it gave me was a blanket over speaker.

Turns out cobra's fizz is largely coming from it's big coupling capacitors. Who knew cutting low end would calrify highs aswell.
I removed R4, and changed C9 to 2n2 and C41 to 3n3. Most of the fizz is gone, even at lower channel- and master volumes. Even the precence is somewhat useable now.
This also solved the need for attenuating fx-send level.
I feel like I'm getting even stronger bass, but it's not overloading the fx-loop anymore.

There is still some fizz on low volume and higher gain levels, but it also gets a bit mushy with higher gain.
So I guess I want the amp to feel a little easier to play with gain at current level. I think it comes down to cathode resistor values..?
Currently I have 1k8/1u, 2k7/1u, 12k, 820r/.68u

[Zozobra]

Cool to hear you've made some progress with it

I guess when you look at what the coupling caps are doing from an intermodulation distortion perspective then lowering them could remove hashiness be restricting the bandwidth contributing to IM. Comparatively, 2n2/3n3 is still quite big when compared to a modern 5150 design or Engl. The first coupler in a 5150ii is 470p! For modern metal tones a hell of a lot of bass is cut early on.

[theghostnow]

So apparently it took me almost a year to solve this issue.

I tried swapping, adding and removing a lot of parts. Some were an improvement, most were not.
I found the smaller cold clipper to add fizz, but I kinda liked the fluidity it gave me so I have it on a switch to go between 12k/17k IIRC.
Changed plate resistors, added plate bypass caps, removed cathode bypass caps, increased nfb etc. and all it gave me was a blanket over speaker.

Turns out cobra's fizz is largely coming from it's big coupling capacitors. Who knew cutting low end would calrify highs aswell.
I removed R4, and changed C9 to 2n2 and C41 to 3n3. Most of the fizz is gone, even at lower channel- and master volumes. Even the precence is somewhat useable now.
This also solved the need for attenuating fx-send level.
I feel like I'm getting even stronger bass, but it's not overloading the fx-loop anymore.

There is still some fizz on low volume and higher gain levels, but it also gets a bit mushy with higher gain.
So I guess I want the amp to feel a little easier to play with gain at current level. I think it comes down to cathode resistor values..?
Currently I have 1k8/1u, 2k7/1u, 12k, 820r/.68u

What value are the cobra coupling caps?

You have to think that early before distortion what the designer choses gets distorted and produces harmonics and then again and again so if some of the fizz is produced in the area of 6 khz who is to say that it is not a harmonic that started from a lower frequency right there between the first and second stage.

A lot of amp guys choose to actually cut bass early to keep it tight and reduce some fizz.

I don't find that necessary I mean palm mutes turn fizz to chunk and as you said it does give some fluidity. The opposite is cut too much and you have a more tight but sterile sound.


Some prefer to reduce the first couing cap and then the second by equal values or the first more and the second half of the reduction.

Some prefer to utilize the first stave bypass cap for that.


But why?

Why commit. You have a small eq pedal or a tube screamer do that exact job without taking away the ability of the amp to rock out more especially with single coils in my opinion.


Now for the fx loop when I had my dragon the best thing was to use a parallel loop which was already there and have one channel for rythm one for soloing.

The parallel config already is a mixer on the framus loop so you can pretty much avoid overloading the effect.

But you have to have an effect that

a) has at least +6db or more capability
b) high enough input impedance at least 10 times the output of the send
c) the ability to send back to the amp only the effected sound to avoid comb filtering by adding two dry sounds (the amps internal and the multi effect one which will be delayed by the ms it takes for the multibeffect to render. Hello phase cancelation).


For me the amp was perfect as it was with a g major. The g major is fully adjustable and can take up to +20 db with input and output control.

You also have the assignable solo boosts...


Also if you want a tighter sound why not applying more negetive feedback but I think you already tried it.


My moto is get away with as much of a loose sound you can instead of the opposite. You can always boost it later on and the more you push the amp the fuller it sounds vs an internally modded one.

[theghostnow]

So it 47n.

Typical rectifier slo values are 22n.

5152 has 0,044 there though but pumps bass later with the ressonance control.


All of that is relative. The framus had a fake depth but I would say my avenger had the most quality control over bass, very feeling but never in the way the rectifier seems to lose control. You can't dime the bass on the rectifier but you can on the soldano. Same preamp. Different choke properties different transformers etc. The framus you could go sub low but I still prefer the soldano over it.

[Moodknyller]

Thought I'd share a little update here, been lurking at sloclone for the past year or so. Turns out I know next to nothing about tube amp technology!

I had a g-major for a long time, swapped it with a DD-500. Hated the parallel loop, it gave me weird sounds and phase issues.
Too much send level distorted the delay, even though DD should have enough headroom for line level.

I put a merlin style loop in my Cobra, with adjustable send and return levels at the back. Threw away the mix pot, because I had no use for it anymore,
and put a soldano-style depth instead of a fixed depth in NFB. So now I have Depth and Deep at the front panel.

Using KT77 now BTW.
First CC is 47nf going to a switch, so pos1 47n->1n, pos2 47n->1n//560K, pos3 47n->1n//1n. After that 3n3 and 4n7. Helps with different guitars and styles of music.
What I noticed recently, was that the grid stopper between 3rd and 4th stage was giving me a lot of hiss and fizz. Too much grid resistance was a blanket and gave even more idle noise.
So I turned it into a voltage divider. It made the amp dead quiet and fizz free with good highs and clarity, but lost distortion and felt like low notes were almost clean, if that makes sense.

So now I'm trying to find a balance having 4n7, 150k, 330k to ground, 68k grid. These are probably stupid values to use, and if someone with better understanding of the maths behind it wants to educate me, please do so.