6G15 Stand Alone Reverb, with Scratch Layout - Finished!

[Willabe]

Well I've finished my scratch layout for the eyelet board. ( I think?)

I did this because I wanted to learn how to use the sch. program and to take it all the way to a finished working product. It's one thing to look at the screen (size/placement wise) and another to print out a template and lay the parts on it.       Took a while to correlate the two.        

I tried to use all I've read on grounding and lead dress from here/Doug, KOC's TUT books, Merlin's books, and Aiken's site.

One thing led to another and this is what i ended up with.        

I'll make the board tonight or tomorrow, and I have all the parts. I've already mounted the iron to the chassis, it's a mojo clone, and I have the mojo cabinet box for it.

The only thing I added to the circuit (besides 2 grid stops , a separate screen filter node/screen R, grid return R for return stage) was more filtering, small radial caps. I did use 1 40uF sprague cap I had laying around. Also went with a FWFB. Oh, and a dc standoff for the heaters.
I feel pretty good about it. I learned a lot doing this. I think it's gonna work fine.


                         Thanks,     Brad      

I changed the 14ga. buss wire to green to make it easier to see. The buss wire is lifted up above the parts below it.
    

[LooseChange]

It's obvious you've been reading O'Connor. Nice job.

[HotBluePlates]

If "PP" means polypropylene, your coupling caps are likely going to need to use a radial type of arrnagement; they're likely gonna be bigger than you allowed for them (unless you measured parts-in-hand prior to your layout).

Looks good. I like it!

[Willabe]

Yes they are polypropylene and they are radials. I measured all parts in hand prior to making the drawing, and laid them on the printed out template to make sure they would fit. Those pp, .01 radials will drop right in without bending the leads on them.

Do you think .01's are large enough or do I need to go bigger? Like .1uF?          I put them in to kill any high freq. hash.  

And thanks to both of you for the good remarks/grades.


              Brad      

[phsyconoodler]

Interesting layout.Far too complicated for this kid.

I've built a couple using this Fender layout.They sound marvelous.

[Willabe]

I've built a couple using this Fender layout.They sound marvelous.

I bet they do. I've seen some of your builds and you do very good work. I've also seen you trouble shoot a good number of things here where you nailed it.

I just want to try out some of these different things I've been reading about and see if I like them. At this point they sound good to me when I read them on paper. But then again maybe there not necessary or worth the extra time, build space/parts count, or money?

I don't like the half wave PS fender used and with the price of the small sized radial caps I can add more filtering, while still at a lower price. Since I'm making my own eyelet board, why not try it?     

I thought this would be a good build to try them on.           

                        Thanks,      Brad              

[HotBluePlates]

Do you think .01's are large enough or do I need to go bigger? Like .1uF?

Well the filter cap bypasses are overkill, but it's your baby and I won't be the one to say she's not as cute as you say... 

It's worth trying and forming your own conclusion.

[zendragon63]

Impressive that you have taken the time effort to layout to proper scale (I don't have quite that much patience and usually end up building/shoehorning around the iron and such).

Though I have tried it on a couple of builds, I can't really hear any of the upper end clarity that some claim adding the smaller bypasses on the e-lytics filters or even the cathode bypass caps. Maybe I just can't hear it above the tinnitus. Looks like a worthy project Willabe--good luck! Regards

dennis

[Willabe]

Thanks HBP and dennis.

I cut the eyelet board last night and taped the templet to it. I installed all the pots, jacks, ect. and layed the board in. Looks good to me so far. Looks like it's going to work out good.


                     Brad             

[Willabe]

All right, I stripped out the E-cap bypass caps.

Here's what I think I'm gonna try.

Please let me know what you think.

( Eyelet board is 14 5/8" x 3" )


          Thanks,     Brad      
      

[sluckey]

don't like the half wave PS fender used and with the price of the small sized radial caps I can add more filtering, while still at a lower price. Since I'm making my own eyelet board, why not try it?

What will your HT winding AC voltage be? If you use a bridge with the same PT that Fender did you'll have a LOT more B+ than the original.

[Willabe]

What will your HT winding AC voltage be? If you use a bridge with the same PT that Fender did you'll have a LOT more B+ than the original.

   
        Oh great,            Thanks for pointing that out sluckey.        Hmmm...             Well, I could go back to the HW Fender used. I do have a lot more filtering in place, should be quiter than the Fender?

Or maybe I could put/add a small droping R, say 200R to 470R, between each of the 3 - 20uF resevoir caps to knock it down. (I do have some R's (10R / 5w) in the PS path already.) It is class A, so it's current draw is pretty much going to stay constant? Do you think that would work?



                Thanks,    Brad

       

[Willabe]

OK, I see in the library that Doug's SA Verb. and the Revibe are both using a FWFB, with a mojo verb PT. Since I have the same mojo verb PT and a MM verb PT (which I was going to use) I'm building a FWFB PS so I can check the voltages unloaded. If there close I should be good to go?

Can I put a R across the output to simulate a load to check the voltage when loaded to make sure it's right? If so what value should I use?


          Thanks,     Brad        

[Willabe]

What will your HT winding AC voltage be? If you use a bridge with the same PT that Fender did you'll have a LOT more B+ than the original.

OK, I built a FWFB dc PS, with the same set up as in the schemo drawing i posted. FWFB, 3 current in rush R's at 10R - 5w each, then to a C-L-C pi- filter, 60uF- Merc. Mag. choke- 60uF, and put it on a variack. Seems to be working fine.

Here's what I'm getting for voltages, unloaded with the variack set for 120Vac.

HT, Red - Red/Yellow, 286Vac

B+ dc out put after C-L-C pi-filter, 362Vdc (no load)

Fender schemo shows 295vdc after pi-filter, full load.

Is it still going to be to high when the full load is on it? Or do you think it will come down enough to get close to Fenders 295Vdc?

I'll try the mojo verb/revibe PT and I'll report back with it's voltages.


                     Thanks,    Brad          

     

[tubeswell]

Hi Brad - Those voltages look like they will work fine. More importantly, how's the hum levels? Pete

[Willabe]

Hi Ya Pete, (how's my good friend in NZ? Hey your headed towards spring! Starting to get cold here.)

Hi Brad - Those voltages look like they will work fine.

That's great to hear, thanks.    

 My heart sank when sluckey asked me what the HT volatge was      (and Oh No!   )   and that using a FWFB might sink my ship.  

More importantly, how's the hum levels?

I only made a small FWFB dc PS on a 4"x 5" eyelet board to test the PT's for voltage's, so I don't know yet. I should have this anyway for test gear. Later I'll put it in a chassis/box and add some barrier strips, banana jack's and test point jack's. I've got the part's here layin around.

Next I'm goning to see what the mojo verb/revibe PT put's out with the same  dc PS board.

I saw you posted the other day, that your working on a combo with sluckey's Warbler in it.        You have very good ideas, can't wait to see it!


           Thanks Pete,   Brad          

 

[Willabe]

All right I took some more readings. All readings taken with Variac set to 120Vac, and the pi-filter when used is CLC (66uF - Merc. Mag. SA verb choke -66uF).

Merc. Mag. stand alone Verb PT , half wave rectifier;

HT = Red/Red Yellow, with nothing hooked up to it, 287Vac.

B+ dc filtered out put, unloaded = 365Vdc.

B+ cd filtered out put, 8K2 load (resistor to ground) = 287Vdc.

Merc. Mag. SA verb PT, FWFB rectifier,

HT, same as above, 287Vac.  

B+ dc filtered out put, unloaded = 365Vdc.

B+ dc filtered out put, 8K2 load (resistor to ground) = 310Vdc.

Mojo/Doug's SA Verb/Revibe PT, FWFB rectifier;

HT, Red/Red, with nothing hooked up to it, 251Vac

Filtered OP, unloaded = 320Vdc.

Filtered OP, 8K2 load R to ground = 285Vdc.

I did not try it with a HW rec. I believe this PT was made for a FWFB, because Doug is using one wired up this way and the voltages look to be right.I don't know if the way I loaded the circuit is an accurate way to measure the voltage drop of a loaded PT? But if it is in the ball park there's only 25Vdc difference between the two FWFB rectifier out put. Might be to high for the screen grid and the MM choke? If so I could break up the 1'st filter cap bank (66uF) into 3 with a small dropping R in between them or would sag/compression be a problem? (Class A, so not a large swing in current draw?)


             Thanks,    Brad                      


  
  
  

[Willabe]

OK, I changed the first bank of filter caps,(from 3 @ 22uF, in to 1 @ 22uf - droping R @ 5w (value ?) - 2@ 22uf) in the PS to add another droping R, just in case I need/want to lower the B+ a little more.

Unless someone see's a problem with it, please let me know.      

Other wise I'm going to try this layout.      


  
                     Thanks,       Brad      

[Willabe]

 I need to bring this up top . 


            Brad     

[tubeswell]

The reason the fender 6G15 voltages are aiming for a B+ around 300 is to run the 6K6 in the ballpark 6-8W dissipation.

If you're going to run a 6V6, then a B+ of 350-360 will work fine (and you can change the cathode resistor to 470R - just like a 5F1 or 5F2A, and run it with an NFB loop from the RT secondary to the cathode of the pre-amp driver)

[tubeswell]

FWIW I added a 39k feedback resistor from the RT secondary to the cathode of the pre-amp driver stage that is driving the pentode and it helps clean up the dirt quite a bit. (I had to rig up an speaker to the RT secondary when I tested this to make sure the NFB wasn't hooked up the wrong way around) I also ran the cathode at 500R with a 6V6 and added a 470R screen grid resistor.  This way I get plate voltage of 338, scr of 341, and cathode of around 18V. Drives the 6V6 nicely.

[Willabe]

Hi Pete,

Sorry for waiting a day to write back, I wanted to think about what your saying and re-read some in KOC's TUT books. The only thing I found was that he said that about drive power for the input tank coil is "the input coils of most reverb tanks are designed for 1 watt or less of electrical power." Accutronics says the tanks that we would use for tube drive need 28mA drive current.

So how many watts output do you think your driving the tank with the way you have it set up? Looking in a couple of tube books, 6K6 can put out as much as 4.5w in SE and 6V6 can do 5.5w SE, (at a little less then the plate/screen voltages Fender and you are using , which is well known because of champ type amps) it seems like too much for the input coil to handel with out burning up.  

What can they realy take before they burn up? I can't seem to find out, I thought Accutronics would have that on their web site , but I don't see it there.

FWIW I added a 39k feedback resistor from the RT secondary to the cathode of the pre-amp driver stage that is driving the pentode and it helps clean up the dirt quite a bit. (I had to rig up an speaker to the RT secondary when I tested this to make sure the NFB wasn't hooked up the wrong way around) I also ran the cathode at 500R with a 6V6 and added a 470R screen grid resistor.  This way I get plate voltage of 338, scr of 341, and cathode of around 18V. Drives the 6V6 nicely.

Thanks for that info, I'll try it.


             Thanks,     Brad          

[tubeswell]

I'm using a SE OT with a 5k reflected load. This is a voltage ratio of 70.7:1.  I checked my cathode resistor and its 600R, and the cathode voltage is 20.3, so the 6V6 is running at 33.83mA. I also have significant attenuation in the pre-amp stages and am running a 12AY7 as well (instead of a 12AT7), so the voltage swing at the 6V6 grid is barely enough to cause it to clip.  Assuming a peak plate output swing of 270 at the most (but it could be quite a bit less than this because of the pre-amp attenuation - I haven't scoped it to actually see), that means the secondary side would be seeing about 3.8V peak (or 2.6V RMS) @ 2.4A which is 6.25W AFAICT,but it could well be less than this.

The end result of the NFB is that the wet signal is quite a bit cleaner (than it otherwise would be if it was the stock circuit). I can turn the dwell full up and it sounds nice'n'wet, but relatively undistorted

I also have a 470R 5W in parallel with the RT secondary to prevent the RT from blowing if the pan input transducer goes open. This also helps keep any oscillations (from the NFB loop) under control. I could probably go a 'notch' lower with the NFB resistor and get the signal smoother still, but I haven't tried it.

[Willabe]

I was worred if I raiesd the B+ that it would end up driving the input coil too hard and burn it up, but your hitting it pretty much full tilt with a 6V6 with no problem. So one less thing to worry about.

KOC is a big fan of putting a resistor across the secondary of the OT, don't remember what value he recomends but it's in the same ball park.

Thanks for all your help Pete.     


                          Brad     

[Willabe]

One more thing, I was planning on raising the pre plate voltages up a bit. 100vdc is pretty low, a little higher should be a little cleaner? Say to 160vdc or so for the input and 180-200vdc for the driver. I've got the PS set up to be able to do this on the 12AT7 independent of each tubes plate. 12AX7 both plates are fed from the same B+ node but it is separate from the 2 12AT7 plate nodes so I can play with this nodes value also.

Where are you small bottle plates sitting?


                Thanks,      Brad       

[tubeswell]

I'll measure it later and get back to you.

[Willabe]

I'll measure it later and get back to you.

No hurry.

   
        Thanks,    Brad       

[Willabe]

Well I finished it and this thing sounds very good to me. The verb is better than on my BFPR and BFSR.

I'd have to say this is perhaps the hardest thing I've done in my life. What I mean by that is the hole thing, from learning how to use ExpressSCH and make the layout to scale, mod the steel chassis, make the eyelet/turret board, wire it up, take pics, learn how to resize them, ect.  

It was a great learning experience and I'm pretty happy with how it turned out. I --- would not --- have been able to do this without this forum and all you guys here! Thanks for all the great ideas/help you all offer, all the time. This is a great place to learn.  

Voltages;

Wall AC= 120.2, filament= 6.75vac, (seems a little high for a clone from MM?) filament DC stand-off = 66.2Vdc

12AT7, V1a; 1=122.2vdc, 2= 0vdc, 3=1.87vdc

           V1b; 6=129.8vdc, 7= 0vdc, 8=2.02vdc

12AX7, V2a; 1= 172.7vdc, 2= 0vdc, 3= 1.35vdc

           V2b; 6= 270.5vdc, 7= 117.5vdc, 8= 131.8vdc

6K6/6V6 (NOS, RCA 6V6 gray plate installed)

Plate,        3= 303.0vdc

Screen,     4= 289.3vdc

Grid,         5= 6.7mvdc

Cathode,   8= 24.15vdc

            
 




                      Brad      

[Willabe]

More pics;

[DummyLoad]

congrat! looks very nice! 

[Willabe]

sluckey had pointed out earlier that I might have a problem with high B+ voltages because of changing the HW rec. to a FWFB. So I added a couple more dropping R's to be able to get it with in reason. I just guessed on all the extra/different dropping R's and it worked out very good, can't believe it.

I think I'm gonna try raising the plate voltages for V1 up to at least 160 vdc and see if it sounds any better? Any thoughts on this from any one?

      
              Thanks,      Brad    

For got to say it's very quite as far as I can tell. I ran it in to my BFPR with my Strat, which is a kind of noisy pair. I don't hear any extra noise. I did run the verb PT output into a 12" speaker and it was dead quite, very clean.

I also did as tubeswell suggested and added NFB, I used a 33K FB resistor. (Thanks tubeswell.) Have not tried it without the NFB.

Can I use a small body SPST switch to turn the FB on/off? I would think most off the voltage/current would go to the tank and very little to FB loop?  



    

[DummyLoad]

you have 3.117KV applied to a 450V cap? 

 

[Willabe]

you have 3.117KV applied to a 450V cap? 

Doh!       

Thanks, I'll fix that.


                 Brad       

[Willabe]

congrat! looks very nice! 

Thanks DL.


           Brad     

[Willabe]

Here's the link to the other thread on this. It has more pics and info. (I never should have made this 2 threads, sorry guys.    )


 http://www.el34world.com/Forum/index.php?topic=12566.msg117132#msg117132



           Thanks,     Brad     

[Willabe]

Well I couldn't raise the small bottles plate voltages. I had given them each a 4K7 plate R to be able to adjust the plate voltage if needed. I tried as small as 470R but could only raise the plate voltage 1 or 2 volts.        I should have known that they weren't drawing enough plate current to adjust them upward.

All in all I love this thing as it sounds great and am very happy with it.

Maybe because the pre amp and driver are run clean they don't need any higher of a plate voltage?    

Would still like to hear if there is any benefit of running the recovery stage at a higher plate voltage, oh well.

I guess you guys didn't like what I built, because only DL said any thing about it when I finished it?   


             Brad        

[HotBluePlates]

I guess you guys didn't like what I built, because only DL said any thing about it when I finished it?   

Hardly! You did a phenomenal job! I hope you're enjoying it!

[Willabe]

Thank you HBP.

I am enjoying it.


              Brad