5E3 Tweed Deluxe Re-Build/with scratch eyelet board layout

[Willabe]

Well, after looking at my 1'st build, I decided to re-work it. I built it 15 (?) years ago and it worked/sounded fine from the get go, but I wanted to try a few new things I've learned since then. It's a built on a clone tweed 5E3 chassis.

So here goes, here's what I went for/did...

1. Wired ground star/buss.(As per our host Doug, KOC and Merlin, etc...)

2. Use radial lead e-caps in the PS (Nichicon VZ(M) series, 105 deg. C, high ripple, 22uF@450vdc, low ERS, under a buck each), 1/3'rd the size and $ compared to Sprague/Atom, IC, etc.

3. Filtered DC stand off (taken/fed from a quieter, latter/down stream B+ node and quieter node, ground current wise) @ 70/80 vdc on the 6.3 filaments. Can be necessary with a CF/PI with a high dcv on the cathode with modern small bottle tubes. (so not to exceed the filament to cathode voltage, which is well known here posted many times, again with modern production tubes, to kill them with in seconds.)  

4. Move coupling caps (where possible), to keep grid wires as short as possible and plate wires long, which are shielded by the B+ on them.. This allows less shielded cable, less high end loss (sparkle), by leaving the plate dcv. to act as a shield.

5. By making my own eyelet/turret board - and - by using the small sized e-caps, it gives much more freedom to keep the individual plate/grid/cathode/B+/- ground star loops much smaller, by being able to place the e-caps very close to where their feeding there own B+/- node.

6. I added - only - 2 B+ e- caps-, downstream from the 5E3's traditional set-up, to set up the wired ground star/decoupling/with their own ground star, with the added benefit of giving a little more B+ filtering down stream for the most sensitive low signal stages.

PLEASE MY TUBE BROTHERS, ALL COMENTS, IDEAS, and CORRECTIONS are ASKED FOR.    

Here's the scheme and layout....



         Brad      

Edit: #4 is wrong on my part, as explained by fellow members (thanks jojo, HBP and PRR) latter in this thread.            
  

[tubeswell]

It seems like all the connections will work. (Its a bit of a brain teaser looking at a different layout from the usual 5E3 layout).

[tubenit]

Did you mean to use 33k input resistors?  I do that IF there is one input resistor. Otherwise I do the 68k & 68k.

If I understand what you've done correctly, then you'll have 33k and 17.5k inputs?

with respect, Tubenit

[sluckey]

and plate wires long, which are shielded by the B+ on them.. This allows less shielded cable, less high end loss (sparkle), by leaving the plate dcv. to act as a shield.

Please tell more about B+ acting as a shield. I've never heard that before.

[Willabe]

Did you mean to use 33k input resistors?  I do that IF there is one input resistor. Otherwise I do the 68k & 68k.

No, I'll fix that, thanks Tubenit.

Please tell more about B+ acting as a shield. I've never heard that before.

That's in G. Webers books. He likes to put the tone stack caps on the tone pots to keep the "grid wires short, plate wires long". He says the dcv sets up a static field around the wire. He also would flip the grid/plate wires on the LTPI on Fenders/Marshalls to shorten the grid wires. A filament dc stand off does the same thing (KOC books, as I understand it). The standing dcv on a CF'ers cathode would do the same. I did this in the stand alone reverb unit I built and had no problems with it.    


              Thanks guys,    Brad        

[Willabe]

OK, here's the fixed schemo's as editable jpegs. My grid spacing = .0625, snap spacing = .03125. Settings are adjusted by going into, View (top bar, left hand corner), than click on options and adjust settings. All parts to scale with these settings.

Now to make the eyelet board.


                  Thanks,    Brad    

[jojokeo]

This is good practicing techniques that will benefit you later on in higher gain amplifiers where some of these ideas actaully are more helpful - some more than others. But, for this build you will not notice much if any difference than a standard stock build from the original amp.

I've got and have read every book by Gerald W and there's many helpful things in there some things that causes you to think (which is a good thing) but there is also ALOT of hype too. It's good reading but take those books with a grain of salt. Everything he says is definately not gospel. He's selling amps, amp camps, etc. as his hype is to get you to "buy-in" to his thinking and the need for his knowledge, amp camps, and buying his products if you can't do things on your own. If you were on the Titanic he'd get you to believe that you need one of his anchors.

[Willabe]

If you were on the Titanic he'd get you to believe that you need one of his anchors.

    Yeah he's pretty good that way.

This is good practicing techniques that will benefit you later on in higher gain amplifiers where some of these ideas actaully are more helpful - some more than others. But, for this build you will not notice much if any difference than a standard stock build from the original amp.

Thanks for your input jojo, your proble right. But thats OK, I'm not looking for a big change, just want to clean it up a little.  

This was my 1'st build, (15 years ago or so   ) and while I have had no problems with it, it was not the best to look at.

It was about time for new PS/K by-pass caps and I don't trust those black fiber eyelet boards, I can make my own fiber glass eyelet boards now. My soldering skills are much better now too. I also wanted a stand by switch, don't need the ground switch with a 3 wire power cord. I'm going to keep the cc plate R's but change the grid stops/grid returns/K R's to metal film. Change to a wired star/buss ground, etc.

All in all, just a lot of little things that were buggin me.         So i'm just gonna do it .    


                      Brad      

[Willabe]

I should have asked you guys if you think that the standing dcv on a plate wire does work as a shield?

The grid is known to need shielding sometimes and I don't recall any one shielding a plate wire.    

Some use a "hot" shield on grid wires, were they put the plate dcv on one end of the shield instead of running one end of the shield to ground.

So what do you guys think?


            Brad      

[jojokeo]

I should have asked you guys if you think that the standing dcv on a plate wire does work as a shield?

No, it's not the same thing as a shield and the dc voltage isn't responsible for achieving quiet operation or interference interruption - it's impedance, signal voltage, and sensitivity is much higher therefore less succeptible. The high impedance low-level signal on the grid's input and the grid wire itself acting as an antennae wire (especially when long) is what's repsonsible for noise and interference. Then couple this with the fact that your signal strength is so low at this point and also gets amplified so many times that anything else riding along w/ it also gets amplified too. This gets carried along through the entire circuit getting reamplified again and again so this is why the input stage is so important, sensitive, vulnerable, and therefore critical to taking care here.

Remember also that if your transformers or filaments are too close they can each induce hum into the circuit here easily as well and other things like ground loops too. This is the most sensitive area of your whole circuit which influences the whole thing. So in your attemps to adhere to the many "rules" of proper layout and design - missing only one thing can ruin all of your other attempts and good practices. The weak link in any system can be the undoing of the whole thing.

There's so many things to understand regarding all of this and how they relate to each other or interreact that it takes most "ordinaly humans" several builds to get right and sometimes even more. Keeping all of this in perspective, lower gain amps are not nearly as sensitive to many of these things.

It's when some ambitious guys try to make the high gain monsters right away before they even realize if they're really actually ready that suffers from the frustrations that occur and that we see here trying to debug many of these kinds of mistakes. But, there's the caveat - if you don't try you'll never get to where you want to go. It's part of the challenge and fun which should cause many to be happy & proud when they achieve success. It's usually just not that easy of a thing to do and do well.

The grid is known to need shielding sometimes and I don't recall any one shielding a plate wire.   

Some use a "hot" shield on grid wires, were they put the plate dcv on one end of the shield instead of running one end of the shield to ground.

Long grid runs especially early in the circuit should always be handled w/ care and shielding is good insurance.
The hot being the shield is one of Weber's "tricks of the ninja" in doing whatever it takes to remedy a parasitic problem when all else fails and should not be a "standard technique" in your building. It is rather unsafe in certain circumstances and is a remedy/response for something not done correctly in the first place in most cases.

Happy re-building Brad -

[HotBluePlates]

Jojo's telling you right.

A shorter way of saying it (I'm learning from PRR, as I'm usually long-winded):

Plate output is lower impedance than grid input. The coupling cap is the delineating point for low/high impedance.

High impedance circuits need to have a small "loop area" to avoid acting like an antenna. Loop area is the space taken from grid wiring, tube and cathode resistor/cap wiring. Low impedance (even if relative here) is more immune to negative effects from noise.

The low impedance at the plate is afforded by the parallel path, at a.c., offered by the tube's internal resistance and the plate load to the filter cap (both paths run to what looks like ground to an a.c. signal).

[PRR]

> Plate output is lower impedance than grid input

Also 20X-50X stronger.

But shields work both ways. They keep stuff from sneaking in, they also keep stuff from sneaking out.

Mostly there's 2 or 3 VERY low-level points in a guitar amp. Input, volume pot wiper, maybe tonestack output. Maybe reverb return. If these are shielded most trouble is averted.

Distance is often an alternative to shielding.

With Fender classic layouts the in-jack wires may be routed at the end far from all else, and often OK without shielding.

The pesky pot-to-grid lines tend to run near trouble. Even a Champ can be ultra fussy about tone/vol return wiring.

The power stage plate leads MUST be kept away from low-level stages. You could run them in shield, but there's no common cable rated for such high voltage. Since the big bottles stand between low-level stages and OT, distance is best.

[Platefire]

Willabe

  Your kind of doing the same thing I'm doing with my old Bogen/Bassman project, except it has been more like 5 years for mine. My amp has always been noisy, cold, un-responsive and I knew at some point I needed to go back in and do some correcting. Meantime it has been sitting up un-used. So I'm trying to apply my more advanced learning and I'm amazed at all the mistakes I found. At the same time of correcting/improving the amp the additional discussion has brought on lot more very usefull info and learning.
  What I wanted to say that this thread discussion is very interesting and usful for me. You definatly left a lasting impression on me for the importance of protecting/shielding the input grid and early preamp gain staging wiring. I've been told over and over especially by tubenit but now I think I finally have an indelable impression on my brain. Thanks, Platefire

[Willabe]

Thanks for set in me straight guys. I appreciate it. I put a note/edit on the 1'st post of were I was off track.

Platefire thanks, but I'm just trying to figure this stuff out myself.   

Got my board built and started stuffing it last night. I'll post pics when I get it done.


           Brad