Please confirm me this treble bleed schema...

[massygt6]

I'm sorry if the argument has been discussed so many times, but from the description founded on the forum I can't recognize the typical schema of a passive low-pass filter, so I draw one by myself, to understand if I've figured it out...
Please, can you take a look at it?
Many many thanks...
[img601]http://img444.imageshack.us/img444/5509 ... eedjg7.jpg[/img]

[dimitris]

I'm afraid that your schem is wrong .Check mdroberts1234 schem in the following link.It has the treble bleed implemented the right way.
Early 80's grail tone

[massygt6]

ops
ok, thank you for pointing me at this scheme, very helpful...

[UltraHookedOnPhonix]

I'm afraid that your schem is wrong

It is not wrong. massygt6's schem is the electronic equivalent of the "grail" schematic. The only difference being that the grail schem has the wiper tied to one of the outer lugs but you can leave one of them disconected and you still have a variable resistor. Also, the only accurate HF taper values I know of come from #123 where a 500K pot goes to a .022uF cap that leads to ground.

Go here and play around with values and make your own filter:
www.muzique.com/schem/filter.htm

[Structo]

Concerning the Hi Lo pass circuits, do you think it makes a difference which component goes to ground?

Meaning should you have the cap or resistor go to ground and what sonic difference would it make on the treble bleed mod?

Is what we are using for the treble bleed circuit a Low Pass filter?
It passes the lows and shunts the highs to ground?

In playing with that calculator, the lower you make the resistor the higher the frequency.

So if I use a .001 cap and want to take the high end fizz out, I should go fairly low on the resistance?

[heisthl]

That calculator if for a high pass or low pass filter, the bleeder circuit is a high cut filter. The calculator only shows you a corner frequency and the bleeder circuit doesn't use that "corner".

[ayan]

Concerning the Hi Lo pass circuits, do you think it makes a difference which component goes to ground?

Meaning should you have the cap or resistor go to ground and what sonic difference would it make on the treble bleed mod?

Is what we are using for the treble bleed circuit a Low Pass filter?
It passes the lows and shunts the highs to ground?

In playing with that calculator, the lower you make the resistor the higher the frequency.

So if I use a .001 cap and want to take the high end fizz out, I should go fairly low on the resistance?

You will hear different responses to the question of whether or not it matters if the cap is tied to ground or the reistor is tied to ground. Personally, I have my own answer and I'm happy with it... and it is: no, it doesn't matter.

The circuit is basically s LPF, snce you're shunting highs to gorund, just like the common guitar tone controls that have existed since before I was born. You should try different values and see what works for you. Grabbing a calculator and trying to see the ferquency knee of the filter will probably not be the best use of your time, since the filter does not follow a buffer. The thing will load down the output of V2B, more or less depending on the the values of the cap and pot. I must admit I never tried the 500K .02uF Dumble filter, that sounds like too much to me... I use a 250K trimmer and a .001uF cap, and the range you get on that is basically from "no load" (almost) when the full 250K is in the circuit, to enough to make the amp sound sound too compressed when the resistor is out of the picture. You can't get the effect you get on a guitar tone control -- where you can go from bright all the way down to "woman tone" -- with my values, the range I get is actually pretty subtle.

Gil

[Structo]

Thanks Gil, do you have any idea what your trimmer is set at?

I built a resistor decade box the other day.

I think I will insert that in my bleeder circuit and see what I get when I try different values.
I guess I could jury rig a pot in there but then I wouldn't be using my new decade box.

I haven't had much luck finding a trimmer pot of the needed value in my town.

The ones I have found were very tiny and only 10K at the largest.

So until I can validate another order online to justify a shipping charge I will make due with a fixed value.

[UltraHookedOnPhonix]

OOPS, Gil was faster than me in answering!

Is what we are using for the treble bleed circuit a Low Pass filter?

That is correct Structo.

It passes the lows and shunts the highs to ground?

Yes, the important thing is that depending on where you set the resistive element it shifts the cutoff frequency.

In playing with that calculator, the lower you make the resistor the higher the frequency.

So say we have a 1K pot set to it's midway point (500ohms) and a .1uF cap, the filter will start rolling off highs at 6db per octave at approximately 3.2kHz. If we set it to the full resistance of 1K, the cutoff frequency is now lowered to 1.2kHz so it is now cutting more highs. The higher the resistance in the lowpass filter, the more highs will be "tapered" off to ground.

should you have the cap or resistor go to ground and what sonic difference would it make on the treble bleed mod?

You most certainly want a cap to ground. If you flip the resistor and cap, you've now made yourself a low-cut filter/high-pass filter. Not cool! Unless that's what you want off course!

So if I use a .001 cap and want to take the high end fizz out, I should go fairly low on the resistance?

Well, if you want maximum treble bleed out of the 500K/.001uF combination, set the pot to 500K and then you'll have a high-end rolloff at 6dB per octave starting at 318Hz. Set the pot to it's midway point and he rolloff will start at 636Hz. Set it to two tenth of it's total resistance (50K) and you'll have a less agressive rolloff starting at 3.2kHz.

Now, look at Dumbles filter which is 500K/.022uF. Maximum treble bleed rolls off everything above 14.5Hz!!! At half pot rotation, you're rolling off at everything above 29Hz (anything below 20Hz is below human hearing range). So you'd have to set it very early in it's rotation to yield the same "spread" as the 500K/.001uF combo.

So, if it's a lot of fizz you want to tame, chuck the D-man filter in there. It starts to rolloff highs at 14.5Hz VS. 500K/.001uF which does so at 318Hz.

Of course, the numbers are only good as a start-off point. From there, let's use the most sensitive instruments (our ears) to figure out the rolloff point!

[ayan]

Thanks Gil, do you have any idea what your trimmer is set at?

I built a resistor decade box the other day.

I think I will insert that in my bleeder circuit and see what I get when I try different values.
I guess I could jury rig a pot in there but then I wouldn't be using my new decade box.

I haven't had much luck finding a trimmer pot of the needed value in my town.

The ones I have found were very tiny and only 10K at the largest.

So until I can validate another order online to justify a shipping charge I will make due with a fixed value.

I set mine at about 125K to ground or thereabouts. Note that my output resistor on V2B is 150K; of you use a 180K one, you may not need the treble bleed at all. I buy my trimmers at Mouser... offered in 500 ohm to 1 Meg range, if I'm not mistaken.

G.

[Structo]

Yes I have the 180K resistor there.

I just have a problem with this high frequency hash or fizz, not sure what to call it.

But it gets annoying after awhile.
I realize we have talked this circuit to death but thanks for going over it again.

Efforts to cure the fizz have been:

Ceramic cap across PI plates. (47pf I think)
Treble bleed (150K - .001uf)

And probably more that I can't recall right now.

330pf treble cap

I do have the 110K/120K plates on the PI.

High preamp plates.

I don't think I changed the mid cap.

150K FET

330R Choke resistor.

470R Screen resistors

3K3 grids

I wish I had a scope and then I could post some traces here and let you smart guys look at it.

[UltraHookedOnPhonix]

Structo

What value are your snubbers? (If non-HRM)

Also, if you bias your tubes cold enough you can get a little bit of a what you're describing i.e. that fizzy sound.

What is your loaded B+? What tubes are you running?

[Structo]

Structo

What value are your snubbers? (If non-HRM)

Also, if you bias your tubes cold enough you can get a little bit of a what you're describing i.e. that fizzy sound.

What is your loaded B+? What tubes are you running?

This is a D'Lite 44, 6L6, 3x12ax7

I have 330pf snubbers on V2

Both V1 and V2 have the 220K/150K plates

3K3/2K2 cathodes, 4.7uf on all except V1b which has 10uf.

110K/120K PI plates and a 4K7 feedback resistor.

15OK FET resistor

6L6's biased at 35ma

Voltage on 6L6's plates 435

So should I bias it hotter and see how it sounds?

As I recall, I thought it was the consensus that these amps liked to be biased on the cold side?

On V2b output I have the 180K that feeds the OD level or ratio.
What would changing that resistor to 150K do?
I have seen that value there.

According to this bias chart I am at the cold end of bias.
With 435 on the plates 50% would be around 36ma and 70% would be around 50ma.
http://members.shaw.ca/house-of-jim/Htm ... ables.html

Thanks

[UltraHookedOnPhonix]

Your idle current for the power tubes is currently at 51%. That’s pretty cold although like you said some guys like Dogears suggests setting these amps cold. For kicks try upping it to 60% idle current or 41mA and see what you think.

Changing the 180K resistor to 150K in your case would give you more volume and a slightly harsher sound. I’d stick with the 180K. Gil and Scott said in a thread somewhere that the vast majority of real-deal non-HRM amps have the 150K resistor there although some have been reported to have the 180K. Like Gil said above, if you already have a 180K resistor there you might not need the HF Taper control at all.

Other than that I’d suggest adding the HF taper control and swapping pre tubes for ones that sound darker. Reportedly the JJ ECC83-S tubes do and as Funk reported, these are the pre tubes that both Robben and Larry are currently running.

[Structo]

OK, thanks for the tips.