trinityamps.com

A higher voltage at the PI cathodes is nothing to be concerned about. It just means a bit less drive is hitting your power tube grids, since the PI's "flloor level" is now around 90V instead of 75V. If you want you could use a 47k PI tail resistor, instead of 56k, to bring it down a little. The traditional Marshall value for this resistor is in fact 47k.

Having said that, this has got nothing to do with any hiss or buzziness problems you may be worried about.

Have a you tried the Zobel filter mod yet? Just make sure to use a non-polarised electrolytic cap for the filter, of the kind used in hi-fi speaker crossovers.

_________________
Great tones come in small glass jars!

Put the design into the Resource section: viewtopic.php?f=12&t=3007

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

On the non-cascaded version, you may want to put V1b bypass cap (C16) back to 47uf or 220uf, whatever it should be, and the V1b coupling cap (C15) back to .022uf. Those components were only changed to tighten the bass when cascading channels.

_________________
Les

http://www.soundclick.com/unimind

Just curious as to which mods, in priority, you would suggest for someonw who wants to tweak, just a little? These would be the key ones that got you closest - fastest.

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

1. Tone stack - C3 and R11
2. V1a Coupling cap - C1
3. Voltage divider - R8, R42 and C20
4. Boost - C20
5. Follower (tail?) - R10
6. Power amp - C8, C9, C10 and C21

Prepare yourself for a ton of gain.

_________________
Les

http://www.soundclick.com/unimind

Interesting. Tone stack first.

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

That was really just the way I did it. The very first clip I posted was the stock Trinity Plexi but with the tone stack mods. Plugged into the TMB channel, turned everything up to 10 except bass on 0. http://www.unimind.us/TrinityPlexi1.mp3

Then I did mods 2 and 3. Then 4 and 5. Then 6. Based on the list I posted above. I think 1,2 and 3 do most of the "brown" tone shaping. 4, 5 and 6 help to refine the tone.

_________________
Les

http://www.soundclick.com/unimind

Oh yeah. Another recent change in the last 2 weeks or so. I changed the PI tail resistor R15 from 470R to 820R like on the sIII and TMB. I think that finally brought the fizz/buzz under control. At least until my over sensitive ears find more fizz.

_________________
Les

http://www.soundclick.com/unimind

820 ohms is the original value used in the 18W PI. 470 ohms is indeed a bit hotter, and also gives some crunchy Plexi-style distortion. Perhaps the extra gain with the 470 ohm resistor is a little too much with EL84s, when the preamp is also cascaded for extra gain. It might be interesting to see what would happen if 6V6s or 6CM6s were used in instead of EL84s, staying with the 470 ohm PI cathode resistor.

_________________
Great tones come in small glass jars!

Edit: See "Plexi brown - fizzie when dimed" thread. viewtopic.php?f=7&t=2978&start=80

_________________
Les

http://www.soundclick.com/unimind

Layout and schem show V1 having 820R and 2K7 cathode resistors but mod notes make no mention of changing one of the 2K7s (stock) to 820R.

The voltage divider cap network has 470K/470K/330pF on the notes but 470K/1M/330pF on the layout.

Notes mention removing R34, one of the boost resistors. Layout and schem still show this resistor.

Also layout shows 100K grid leak resistors on the power tubes. I see no notes of this on the mod list.

_________________
http://pdfelectronics.com

I Had put in the 2k7 to try and reduce a little front end gain to help with the fizzie problem. Although it helped a little, I have recently gone back to the 820r.

Again the 1M is to help reduce gain. It should be fine with the 470k.

Not sure what notes you are referring to but I have removed it and just have an 820R or maybe 1.2K (can't remember) for R9. My boost switch is now a "FAT" switch where it switches between .68uf and 330uf at C20.

Another gain reducer to help with fizzie problem. These brown mods produced a lot of gain on the front end and were slamming the el84s. If you do not have high output PUPs you should be fine with the stock values of 470K. Although I would maybe go down a little like to 330K because you will be driving the 84s harder.

_________________
Les

http://www.soundclick.com/unimind

Hi Les

I am going to use the stock 820R then.


I will go with the 470K, it's 3dB more gain than the 1M. I will reduce gain later.


Ok I will just go with a 2K7 resistor and a switch to take in/out in a 0.68uF cap.


This amp will be used with a variety of guitars so I will go with the 100K grid leaks. I am also including the cascade switch so I think I will have more than enough gain!


Thanks!

_________________
http://pdfelectronics.com

Yeah with all the gain reducing tweaks, I would start with the stock values and then reduce gain later if needed.


Now the thing is if you still want the boost switch then you should have the .68uf across R34. If not then you will not hear much difference between boost or off. Or if you only go with a single resistor and just want to switch the .68uf cap in/out for the boost that will work as well.

_________________
Les

http://www.soundclick.com/unimind

I think 330k should be good here. Some of Richie Hall's higher gain 18W versions also use 330k. For those who don't know, Richie (aka "Plexi" on 18watt.com) was one of the main people who made the Marshall 18W revival happen.

_________________
Great tones come in small glass jars!

While I was building the Plexi Brown I was worried about using 100K grid leaks on the output tubes. I thought it would kill too much gain.

Turns out the output impedance of each tube in the LTP is about 38.5K as I just calculated it. So the 38.5K making a voltage divider with the grid leaks is what attenuates gain. So turns out the different between 470K grid leaks and 100K grid leaks is like 2dB or so. The 100K probably subtlety smooths things out by lower gain.

Either way the Plexi Brown does have a bunch of gain!

_________________
http://pdfelectronics.com

Would the 1M grid leaks and 470/10k cathode/tail resistors be better suited to 6V6s?

_________________
http://pdfelectronics.com

My tube manual specs max 500k grid leaks for cathode bias.

Good point!

Wonder if the 10K PI resistor would drive the output tubes too hard, 6V6 that is.

_________________
http://pdfelectronics.com

Ok I have most of the parts for the brown mod. Still trying to pick a starting point

One question about r9 and the boost switch. In the layout for the plexi brown it seems that the .68 by pass cap is over the secondary 2.7k that's add to reduce the resistance to ~ 1.35 k. How about putting the bypass cap right across r9 always and then adding the second 2.7k as usual . Much of a difference ?