Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: shooter on December 11, 2016, 01:07:46 pm

Title: circuit understanding?
Post by: shooter on December 11, 2016, 01:07:46 pm
Trying to sus-out C10s effect in the R38/R17 voltage divider just prior to the TS.

If I didn’t slip a zero; Xc for 100hz ~ 16kohms,  Xc for 4hz ~ 4kohms

So is this circuit R37/R17/C10 like a variable voltage divider based on frequency?

Title: Re: circuit understanding?
Post by: PRR on December 11, 2016, 01:21:04 pm
> If I didn’t slip a zero; Xc for 100hz ~ 16kohms,  Xc for 4hz ~ 4kohms

Slipped three zeros. Xc for 4Khz ~ 4kohms

So plot the frequency response, assuming that Xc at each frequency works approximately the same as a resistor of the same size.
Title: Re: circuit understanding?
Post by: shooter on December 11, 2016, 04:24:52 pm
Quote
4Khz
:think1:
ya that's what I meant, been a long day of moving snow from where God put it  :laugh:

Quote
So plot the frequency response
I'll wait til my fingers thaw :cussing:

thanks PRR

Title: Re: circuit understanding?
Post by: shooter on December 11, 2016, 07:38:06 pm
Ok, that hurt my prefrontal cortex!

Here’s what I came up with;
Without the cap
Vout = R17 / (R38+R17) * Vin
I picked 1V for Vin, got .5 out, I’m happy

Inserted cap, now
Vout=(R17 +Xc of C10) / (R38 + (R17 +Xc of C10) *Vin, still 1V

At   100hz I get ~ .8V
At 4000hz I get ~ .5V

I’m not looking for a gold star, but would like to know the formula is applied correctly.
thanks
Title: Re: circuit understanding?
Post by: Willabe on December 11, 2016, 09:13:11 pm
ya that's what I meant, been a long day of moving snow from where God put it  :laugh:

If God put it there then you better hurry up and put it back. I left ours where it fell.   :huh:     :l2:
Title: Re: circuit understanding?
Post by: PRR on December 11, 2016, 10:52:36 pm
Xc of 0.01uFd -- now you have me confused. I recall that 0.01uFd is 1Kohm at 16.7KHz (good radio-filter for low-Z lines.

So then I would scribble a cheat-sheet:

0.01u at 40Hz ~ 400K
0.01u at 100Hz ~ 160K
0.01u at 160Hz ~ 100K
0.01u at 400Hz ~ 40K
0.01u at 1KHz ~ 16K
0.01u at 1.6KHz ~ 10K
0.01u at 4KHz ~ 4K
0.01u at 16KHz ~ 1K

Now figure the R17+C10 leg:

at 40Hz ~  43K+400K = 443K
at 100Hz ~ 43K+160K = 203K
at 160Hz ~ 43K+100K = 143K
at 400Hz ~ 43K+40K  =  83K
at 1KHz ~  43K+16K  =  59K
at 1.6KHz~ 43K+10K  =  53K
at 4KHz ~  43K+4K   =  47K
at 16KHz ~ 43K+1K   =  44K

Now, R17+C10 sucks on a cathode follower (~1K) plus R38 43K. What is the effect?

at 40Hz ~  44K sucked by 443K is hardly any suckage. Say 90% or -1dB.
at 160Hz ~ 44K sucked by 143K is about 3/4, about -3dB.
at 1KHz ~  44K sucked by 59K is significant, not-quite half
at 4KHz ~  44K sucked by 47K is very nearly half, -6dB.

Flat below 100Hz. Falling from 100Hz to 1KHz. 1KHz and up, output is half.

Did you know? If you had a signal generator, an audio voltmeter, and a box of R and C, you could mock-up and measure faster than you could figure?

Having got that gold star, you must ALSO consider the effect of the tone-stack. At a glance, it is caps in all paths so the subsonic impedance is infinity. Some extreme-turn says with R3 full-down it must go to R10 47K at some high frequency. if R3 is full, it still goes down to 72K. In fact looking at the above, it is liable to be above 200K in bass but near 47K from 1KHz on up.

I have a real suspicion R38 R17 C10 are redundant. R17 C10 at best stabilizes the impedance of the tone-stack, but to what end? R38 actually reduces both level and maximum boosts. I think. It would take a mock-up or a long hunching over the PC to prove.

My old gut says Leo didn't use those extra parts, so maybe you do not need it.
Title: Re: circuit understanding?
Post by: shooter on December 12, 2016, 08:46:00 am
Quote
Now, R17+C10 sucks on a cathode follower (~1K) plus R38 43K. What is the effect?
I think my brain will overload adding that fact and accompanying math!

I do like "odd" bits of schematics and trying to understand why the engineer? put it there.
I also think it looks like a good?/other way to tame treble.

Quote
If you had a signal generator, an audio voltmeter, and a box of R and C,
They are laying all over my bench, I needed the travel case they were in so.... :laugh:

Quote
If God put it there
I agree, but my wife, um, well she has 1 more vote than I have :icon_biggrin:
Title: Re: circuit understanding?
Post by: shooter on December 13, 2016, 09:10:54 pm
Ok, getting ready to open my last amp and go to rev 1.5
I stuck in the above circuit into my TS.

Did I put it on the correct side of the .01 coupler?
How bad is this gonna suck down my overall signal?

thanks
Title: Re: circuit understanding?
Post by: PRR on December 13, 2016, 09:13:28 pm
> How bad is this gonna suck down my overall signal?

43K+43K is half, but only in the highs.

The James is 10:1.

So 10:1 for lows, 20:1 for highs.

Suckage is NOT an issue if you have gain in reserve. That's what tubes are for.
Title: Re: circuit understanding?
Post by: shooter on December 14, 2016, 08:18:52 am
Quote
but only in the highs.
and that what I hope for, I don't want to change tone, already best I've done so far, just knock it back a bit.

thanks
Title: Re: circuit understanding?
Post by: shooter on December 16, 2016, 06:06:12 pm
ok, I spent a couple hours boosting plate volts and re-biasing my6V6PSE, I also changed the .01 into the TS to .02, then I was wondering...

can I leave out R38 and still get similar results of attenuating highs n passing lows?  my brain says it's still a low pass filter, but what do I know, it's to cold to think!

thanks again
Title: Re: circuit understanding?
Post by: jim on December 16, 2016, 08:15:31 pm
Now there  is some wise words for all electric geetar players:

"Suckage is NOT an issue if you have gain in reserve"  Jim
Title: Re: circuit understanding?
Post by: PRR on December 16, 2016, 08:30:23 pm
> can I leave out R38

With R38, the high-cut cap works against 43K.

Without R38, the high-cut cap works against the cathode follower. That impedance is unsure, but likely near 1K, and if you load it down the CF will distort bad.
Title: Re: circuit understanding?
Post by: shooter on December 16, 2016, 08:45:04 pm
Quote
Now there  is some wise words for all electric geetar players:
I have to say PRR has a mastery of wordcraft!

Quote
and if you load it down the CF will distort bad.
my source is a typical plate coupled, I wasn't sure, but the iron was hot, so
then I did a before/after;
Pre out @80hz      9.3      8.9
@800                   11.5    8.9
@4K                     11.5    8.8
TS centered  all readings Vac rms

then I played :icon_biggrin:
it's great when it works as intended.  The waveforms looked the same before/after
The bass came alive when you attack the strings and stayed chimey sparkly on the other E :laugh:
trying 2 X12 in the morn

thanks for the guidance

Title: Re: circuit understanding?
Post by: PRR on December 16, 2016, 10:23:07 pm
> my source is a typical plate

That is significantly different from the circuit with the R38 in it, no? 12AX7 plate is likely to be near 40K, opposed to ~~1K for a cathode follower drive.
Title: Re: circuit understanding?
Post by: shooter on December 17, 2016, 08:45:11 am
sorry for the confusion, I stole the idea from a CF based TS driver circuit, post 1, but I wanted to implement it in my built which is not CF, post 7.
in the end all things relative seemed to work :laugh:
I need to do some more testing, It works, but I seem to detect high notes decaying faster, i'll scope it after work and compare with before scope shots.