gain - bypass ?

[shooter]

wondering if you split up Rk say from 1.5k to 1k and 470, only bypass one R, would that allow for *some* gain adjustment, while keeping the freq roll-off about what it would be if you just used 1Rk?

[Fresh_Start]

Yes. I did exactly that with a switch from partial to full bypass on one channel of a Super Reverb variant. My objective was to have a lower gain harmonica mode and it worked.

Chip

[shooter]

lower gain harmonica mode

Thanks, I don't want to change the bias point set by Rk, or freq roll-off, just how much gain *boost* you get from cap bypassing. 

[HotBluePlates]

... I don't want to change the ... freq roll-off ...

The exact frequency roll-off will change slightly, because the resistance the cap is working against has changed (the R bypassed by the cap has gotten smaller). You'll probably be close enough with the old value though.

I could tell you how to calculate it, but would you know based on a change in Hz alone what the sonic impact is? Instead, you may wish to simply use a pot to replace Rk temporarily, attach you cap from wiper to ground, and give it a listen. You'll know right away whether you like the sound, and can dial in the exact gain you want for that stage in real-world terms.

[shooter]

use a pot to replace Rk temporarily

my thought is not to change Rk(total), just split Rk = R1+R2(bypassed). As a way to "adjust" a gain-stage without, adjusting - adding, a voltage divider after the gain stage

[clyde]

I've done this several times and find it's a fun process tweaking by ear. 

[HotBluePlates]

use a pot to replace Rk temporarily

my thought is not to change Rk(total), just split Rk = R1+R2(bypassed). ...

Imagine a 1kΩ linear pot in series with a 500Ω (or so) resistor. Perhaps the resistor to tube cathode & "top" of the pot, bottom of the pot to ground, wiper connects to your bypass cap +, and bypass cap - to ground.

If you know exactly how much gain you need, this may be pointless. Then again, it's a fairly long equation to calculate the gain due to a partially-bypassed cathode resistor. The pot/resistor/cap is much faster, assuming you have the parts on hand.

Otherwise, look at your proposed circuit: The resistance to use for a -3dB calculation is the resistance of R2 in parallel with [R1+(Rp+ra)/(Mu+1)].
R1 = unbypassed portion of the cathode resistance
Rp = Plate load resistor
ra = internal plate resistance, at the operating point
Mu = Tube amplification factor

The end result of the equation will be a resistance somewhat smaller than R2 itself. Which also means it's smaller than the original total value of Rk, and with the same bypass cap the -3dB point will be higher by some amount.

How much? Don't know unless you calculate it. How much matters? I don't really know, but I'll know it if I hear it. Which is the real reason I suggested finding the answer to your question empirically...

[shooter]

finding the answer to your question empirically...

Thanks HBP, that's what I'll be doing, I designed this last board with  *jumpers* in the cathode path, Rk is 2.4k, Ck varies depending on which stage, 2.2, 10,22, 22.  I was gonna; || Rk for 1.2k, remove jumper and insert something close to 1.2k, IF, I needed to knock some gain outta that stage(s).
I did clip the math, trying to add all the cool formulas into an Excel sheet of tube math.
 

[HotBluePlates]

... I did clip the math, trying to add all the cool formulas into an Excel sheet of tube math.

RDH4 is available in the Library of Information (link below this thread). It has all the formulae you'll probably ever need. It also often presents some useful approximations.

Don't read it front-to-back (though I've done that). When you have a question about a topic, look it up through the TOC or the Index, and go get your answer. There are a few topics which have nuggets scattered in various places.

The only downside of RDH4 is you really could use a "translator" to explain how/when to use some of the information. But it was intended as a handbook to amass things you already learned/knew, with copious references to other places (now long out of print) where you could get "the rest of the story."

[shooter]

 

Don't read it front-to-back

I just spent the last hour *reading* an RCA manual the PRR linked to awhile back, 40pages of info and 200more of graphs n stuff

RDH4

I'll pick that up next
unless the weather breaks
Thanks for all the help

[sluckey]

Now why not spend 5 minutes doing what you propose. 

[kagliostro]

..... As a way to "adjust" a gain-stage without, adjusting - adding, a voltage divider after the gain stage

A way to obtain that is to split the plate resistor and take the signal from the junction of the two resistors



---

BTW this is how Merlin do the Boost Adjust



Ciao

Franco

[chocopower]

Orange uses a similar concept, but with a pot.


They use a smaller cap so is more a presente control, but the idea is the same.


http://www.drtube.com/schematics/orange/grmk2mus.gif

[2deaf]

When the gain is reduced by placing the bypass cap. in a voltage divider, the relative effect of the roll-off is also reduced.  So if you were getting maybe 2db of cut at 82Hz with the cap. at the cathode, you might only get 1db of cut with the cap. in the voltage divider.  Even if the f_c were the same in both cases, the divider case would sound like a lower f_c.   

[shooter]

Now why not spend 5 minutes doing what you propose

Cuz it's friggin cold and I can't get the chassis prepped - can I move in with you for a week

Honestly I hope NOT to have to do anything, just preppin in case, my bug-out truck is time tested

[sluckey]

Come on down. It's garden plantin' time. We'll drink a beer or two. 

Hope you have A/C in that Dodge!

[shooter]

Hope you have A/C in that Dodge

Up north we call it 2 40 AC, 2 windows, 40mph

[sluckey]

Ooh! You better take a Greyhound then. That 2 40 AC don't work so good on the Gulf coast. Even when the temp is mild, the humidity is still near 100%.

[shooter]

2 40 AC don't work so good

I do have a frig in back, just plumb some ducking, electricity is free
maybe i'll wait til febuary, then I can sit in my short sleeves and watch you freezing

[HotBluePlates]

..... As a way to "adjust" a gain-stage without, adjusting - adding, a voltage divider after the gain stage

A way to obtain that is to split the plate resistor and take the signal from the junction of the two resistors

This came to mind for me, too.

Then I thought, if this reduced gain stage is intended for hi-fi use then reducing gain through local feedback (partially-bypassed cathode resistor) will also help linearize the gain stage somewhat. If signal output only is reduced through a tapped plate load resistance, the actual gain and distortion of the stage hasn't been reduced, only the output voltage.

The lesson here is the intended application matters in deciding how to accomplish the task.

[shooter]

will also help linearize the gain stage somewhat

Thanks HBP, I just read that....somewhere, last night (aikens, valve wizrd, rca )
 btw, I like PRR's RCA book, better than your RDH4, way more pics in the RCA
I did make it through 1 chapter....so far

I thought the same as this is "FI-ish", if I can reduce distortion some, and just put a pedal before the amp to get it back for guitar

[mresistor]

got a link to the RCA book?

[HotBluePlates]

Shooter may be using a paper copy or have different links.

However, follow this page and scroll down to the "RCA" section. Download RC-19 or a later manual; there are sample circuits in a section towards the back. I'm guessing this is Shooter's reading material (though there is basic tube circuit info towards the front that was copied year after year).

[shooter]

RC-19

That's the one, btw, thanks for the link

[mresistor]

thank you for the link(s) sir