Internal Low-pass ... help !

[IslandPink]

Folks
This is proving to be a bit more of a problem than I thought .
I'm just trying to imagine a first-order low-pass filter incorporated into a valve amp ( or pre-amp ) to drive the bass section of the 'Something Stirs in the Undergrowth' speaker combo . 750-ish Hz is the crossover, but getting the exact value is not the issue.
I keep hitting snags when I try to use an RC or LC or LR filter in various places.
Romy does this :
http://www.goodsoundclub.com/Forums/Sho ... ostID=6696
on his Melquadies amp .
If I have a volume pot on the input ( as I do ) then this buggers the crossover point because of the series resistance varying ( I think ) .
In between the stages of an SE amp, I'm worried about how the cap to ground and the series resistance will interact with the grid resistor for the 300B ., not to mention how it affects the anode resistor of the pentode, which is definitely part of the picture.
In some ways it's like a split RIAA phono amp I suppose .

There must be a way to do this ....

ILPS

[Nick]

Yep, I see what you mean, You also want to use a constant impedance filter if possible (just thought about this) as a simple RC or RL will alter the loading on the driving stage, so not only will the HF be attenuated, so will the distortion of the driver stage as the load drops. Now that may not be a big deal as any distortion produces will be attenuated in the LP slope of the filter.

The simplest way I can think is a inductor in series with a coupling cap or interstage between stages.

[IslandPink]

I'll have a think and get back to you in next couple of days .
Keep mulling over this if you can .
Happy New Year anyway, I have to go out to friends now for some drinks
Cheers !

MJ

[Mike H]

It's a right drag innit?

I had to stop soldering last night and go and have "nibbles" (things on cocktail sticks) with the rest of the family.

I'll be glad when we can get back to normal.

FWIW, going back to old circuits with passive tone controls following a volume pot, the TC impedances tended to be larger than the vol pot value IIRC.

If the filter is going straight into a valve grid you can do that.



 

[IslandPink]

If you have any circuits you can link that are relevant, Mike, please do. I'm sure there are solutions to this problem - most things in electronics have already been thought of .

MJ

[Mike H]

Well I dunno we're talking "consumer quality" record players and suchlike!



 

[IslandPink]

I was just thinking of the filtering method Mike. Doesn't matter what the application was !

In the meantime, I think Romy has offered one obvious ( it's obvious now ) solution :
http://www.goodsoundclub.com/Forums/Sho ... 5334#15334

ie. put the filter before the vol pot. Clearly this loads-down the phono output stage heavily at or beyond the crossover, but that's going to happen somewhere in the chain, whatever you do - it's an inevitable trade-off ; but the aim is to get passive parts out of the amp/speaker interface .

I'll report back

Mark

[Nick]

I was just thinking of the filtering method Mike. Doesn't matter what the application was !

In the meantime, I think Romy has offered one obvious ( it's obvious now ) solution :
http://www.goodsoundclub.com/Forums/Sho ... 5334#15334

ie. put the filter before the vol pot. Clearly this loads-down the phono output stage heavily at or beyond the crossover, but that's going to happen somewhere in the chain, whatever you do - it's an inevitable trade-off ; but the aim is to get passive parts out of the amp/speaker interface .

I'll report back

Mark

If you did want constant impedance then I think a Bridged T would do the job

http://en.wikipedia.org/wiki/Zobel_network

[Mike H]

Ah yes keep forgetting how many different "Zobels" there are, in front of the vol pot then? Might be a bit of a prob getting an inductor big enough if a reasonably high impedance is needed


Just remembered there are of course, dare I mention it, op-amp type filters

http://www.educypedia.be/electronics/el ... filter.htm


 

[IslandPink]

Thanks Nick for that info - I'll have to ask you more about that later (eg. where does the pot go in that ? )

Thanks Mike, good database there.

Meanwhile of course the pitfall of the filter before the volume pot is that the outputZ of the phono amp becomes some or all of the series-R in the RC filter ... Damn !! Then if you want to swap to CD you're buggered .

Hmmm & double hmmmmm

[IslandPink]

.. OK, so then I could make the series R much higher ( and cap smaller ) so it would be eg. 10x the Zout of the phono . The pentode doesn't need low-impedance grid drive, after all .... but then I would need to understand how the 100k pot interacts with this ....

However , having thought a bit more about a filter in the space between the C3m and the 300B , it's easier than I first thought - actually I think I'm quids-in by having the pentode driver - the Zout of the C3m is almost exactly the anode resistor ( ~11k ) in which case I can put a cap to ground ( ~0.02uF ) just before the coupling cap ( 0.47u ) to the 300B . How does this sound ?

ILPS

[Nick]

Yep, I meant to use the bridged T filter between stages, its constant impedance would mean the driver valve sees a constant load (LCR RIAA like, the filter shape is the same).