5B4 Circuit build. Resistor question.

[olgoat]

I am starting a build for a Fender 5B4 circuit.  I had another thread a while back and did not want to resurrect it for the immediate question. 


The layout calls out 250K resistors.  It appears that only 220K or 270K are available now a days.   This is the 5C4 layout    https://app.box.com/s/11eqsvwvqhoqv3buhyep
(the 5B4 one is almost impossible to read for the resistor values.).  I am not sure whether to error on the low side or the high side for this particular resister.


There are several other difference between the 5B4 and 5C4 mostly capacitor values and number of caps, but the 250K resistor and position in the circuit is the same.


Any suggestions would be very much appreciated.


Thanks
Tim

[Jack_Hester]

I am starting a build for a Fender 5B4 circuit.  I had another thread a while back and did not want to resurrect it for the immediate question. 


The layout calls out 250K resistors.  It appears that only 220K or 270K are available now a days.   This is the 5C4 layout    https://app.box.com/s/11eqsvwvqhoqv3buhyep
(the 5B4 one is almost impossible to read for the resistor values.).  I am not sure whether to error on the low side or the high side for this particular resister.


There are several other difference between the 5B4 and 5C4 mostly capacitor values and number of caps, but the 250K resistor and position in the circuit is the same.


Any suggestions would be very much appreciated.


Thanks
Tim

Tim -

Here are some 250K choices:

http://www.mouser.com/Passive-Components/Resistors/Through-Hole-Resistors/_/N-7h7z5?P=1yzekiqZ1z0z819

You're right about the layout being hard to read, for some of the values.  Maybe someone has located a better one.

Jack

[eleventeen]

You could use a 470K in parallel with a 510K


result =   product / sum = 244.6K which is only a tiny bit over 2% off.


Or, 510K in para with another 510K = 255K


Either way Leo would say "ship it!"

[HotBluePlates]

Either 220kΩ or 270kΩ would be fine. Each would approximate a 250kΩ 10% resistor which is at an extreme end of its acceptable tolerance.

[PRR]

5C4 schematic is mostly the same, and Doug's copy is easy to read:
http://el34world.com/charts/Schematics/files/fender/Fender_SUPER_5C4.pdf

The critical part is the 250K:6.8K divider. The absolute values are not fussy, but that ratio needs to be near-right to keep the phase inverter balance.

However 270K leads to 7.3K and 220K leads to 6K.

This is one problem where a slide-rule RULES.

330K and 9K work out. 390K and 10K is near-enuff. 180K and 4.7K is very close.

Ah... today Doug is out of stock on 220K. So let's pencil 180K:4.7K.

Even with the lower resistors, the 0.05u caps are good to 20Hz, so don't need changing.

Whatever you use for the grid resistors, you should probably use for the plate resistor feeding them.

Also if you use common 12AX7 instead of now-rare 6SC7, the lower resistors suits the slightly better conductance of the AX over the SC.

[olgoat]

Thanks all. 


I will try and understand your insights.  I know I don't understand why one divider value is better than another ie why 9K, 10K or 4.7K might be ok but 7.3K and 6K might not, but I have seen a number of PRR's posts and if he doesn't know what he is talking about, he sure has me fooled.   


I am unfortunately a solder by numbers guy who got lucky on his previous 5E3 build.  Getting luck twice is not in my nature. 


The sound of the Octal preamp tube is a big part of my reason for wanting to build this circuit.  I am concerned about the 6SC7 as it is not terribly available but I have collected enough of them to give it a go.

[olgoat]

I don't know why I didn't think of mouser to begin with.  For the rank amateur they can be a bit daunting to order from.  (too many choices, which for the rest of you is no doubt a good thing :) but went ahead and ordered the 250Ks from them.  Thanks for the tip.   I have been using 1watt for all the other resisters except the 10watt wire wound. 3/4 was the highest value they had so, I went and ordered those. 


It will be a miracle if this thing runs.    This amp was my goal when I started. The previous 5E3 was just to take the well documented path first to see if I could actually build a running amp.


You guys are a great and I really do appreciate your insights and efforts.   


This was the first time I used Dougs board build software which was great.  Should have gone with eyelets instead of turrets.  Not enough room in the turret holes for multiple wires.  My design is a bit too crowded but I hope it will work.


 

[olgoat]

[PRR]

> why one divider value is better than another

Look at how this phase inverter works.

We need two equal-but-opposite signals to drive the two power tubes.

Top half of 6SC7 drives the top 6L6 directly.

Now we need the same signal but opposite phase.

The other half of the 6SC7 will invert signal. But it also adds gain. We want the inversion, not the gain.

So we cut-down the signal, then gain it back up.

It looks like the designer expected a gain of 37. So the 250K:6.8K divides-down by 37, the 6SC7 gains-up by 37. Result is same signal level but opposite phase. Which is what we want.

If you were forced to dink with the 250K, to 220K or 270K, then you should dink the 6.8k in the same proportion. Otherwise you get un-equal signals, one 6L6 works harder than the other and clips early. Not true balanced push-pull action.

If you worshiped meters and numbers, you would make the "6.8K" *variable* and trim it on the bench for perfect balance. The tube gain could be 35 or 40, but you could trim loss+gain to give exactly equal outputs.

My thinking is that any stray un-balance adds a small touch of "flavor". And that such "flaws" are maybe why you like these older amps best.

[olgoat]

Thanks for the schooling. PRR Much appreciated.