Capacitor Question

[VinylSavor]

In a simple first order high pass filter my understanding is that only a capacitor is needed because the input impedance of the amplifier functions as the resistive load.


You are totally correct about this!

Yes. But do you know the tolerance of the input resistor? This is a non critical value and the manufacturer might not have chosen tight tolerances here as it is totally uncritical. So if you want to have exact filter value you need to take care of this resistor as well. Or you can easily wire one in parallel (doesn't even have to be inside the amp) to take care of this. This way it will not even be in the signal path. This is a super easy problem wich can be solved in less time as it takes to reads this thread ;-)

 

[christensenleif@msn.com]

In a tube amp you need only change first couple cappacitor for input tube or add when not there.If I remember correctly the cutoff frequency was 200hz.
47kohms input needs 0.015uf cap ~220hz

Will still not be very much down an octave under/100hz

 

[Musica Bella]

I believe you may be overthinking this. You have to look at the whole picture. Do the other components in this piece match your goals?Your dealing with a simple first order high pass filter at frequencies where a slight mismatch would not even be noticed sonically. I would get the circuit where you want/need it first. Then work on the enhancements you see fit. Design quality before parts quality.

 

[DasguteOhr]

Will still not be very much down an octave under/100hz

But it's enough to free the amplifier from the main bass load. And Ron can be use his bass tower. You can also set the frequency higher.

Kalkulator für passive RC-, RL-, LC-Filter | DigiKey Electronics

Berechnen Sie die -3dB-Grenzfrequenz von RC-, RL- und LC-Schaltkreisen für passive Tief- und Hochpassfilter mit dem Kalkulator für passive Filter von DigiKey.

www.digikey.de

 

[microstrip]

Yes. But do you know the tolerance of the input resistor? This is a non critical value and the manufacturer might not have chosen tight tolerances here as it is totally uncritical. So if you want to have exact filter value you need to take care of this resistor as well. Or you can easily wire one in parallel (doesn't even have to be inside the amp) to take care of this. This way it will not even be in the signal path. This is a super easy problem wich can be solved in less time as it takes to reads this thread ;-)

You still have to measure the existing resistor if you wire another in parallel. This was discussed in a parallel thread. As you say, and easily confirmed reading the measurements section in Stereophile, most manufacturers only give an approximate value of input impedance.

Wiring a calculated much higher resistor in parallel to decrease the resistance value can be a possibility if the capacitor is chosen to have the closer higher value - 1% resistors are trivial, 1% capacitors are hard to get. But meters are needed, there are no subjective ways of doing it!

RxC must keep the same value, if C is larger we need a smaller R. Adding a parallel resistor decreases R.

 

[christensenleif@msn.com]

WE solved this by winding limited bandwidth output transformers…..
Clever

 

[DasguteOhr]

WE solved this by winding limited bandwidth output transformers…..
Clever

I'd try something like this before converting a 60k amp, just to make sure I liked the result.

ADDING A SUBWOOFER AMPLIFIER

Subwoofer crossover

www.hlabs.com

Or this

FMOD COLOR CODE

FMOD color chart

www.hlabs.com

 

[Ron Resnick]

I'd try something like this before converting a 60k amp, just to make sure I liked the result.

FMOD COLOR CODE

FMOD color chart

www.hlabs.com

These are exactly what I've been playing with the last few months as my experimental step. Thank you.

 

[DasguteOhr]

These are exactly what I've been playing with the last few months as my experimental step. Thank you.

Once you've found the right frequency after experimenting and your amp's warranty has expired, I recommend going to your local radio/audio repair shop and having them install a suitable coupling capacitor for thee first input tube. Then let him measure the frequency response of the amp to see if it matches the corner frequency of your bass tower. Look for an old hand who knows about tubes, he can do it very quickly and complicated. second advantage you can use dc coupled preamps and sources too.
Exsample mastersound

 

[Ron Resnick]

Once you've found the right frequency after experimenting and your amp's warranty has expired, I recommend going to your local radio/audio repair shop and having them install a suitable coupling capacitor for thee first input tube. Then let him measure the frequency response of the amp to see if it matches the corner frequency of your bass tower. Look for an old hand who knows about tubes, he can do it very quickly and complicated. second advantage you can use dc coupled preamps and sources too.
Exsample mastersoundView attachment 153800

Thank you, but too complicated and unnecessary.

 

[Kcin]

Ron this is quite easy and your are correct it needn't be over complex.

I tried to articulate it here- your are applying a simple 1st order filter to the input of the amp:

Capacitor Question

I tried to custom order 2,700pF and 3,000pF capacitors from Duelund, from its CAST copper line, for a DIY high-pass filter. Duelund says it cannot guarantee each capacitor any closer than 20% to the actual value requested. In theory I could have two capacitors that are 40% apart in value! I...

www.whatsbestforum.com

This is not an active filter like @DasguteOhr suggests here:

https://www.whatsbestforum.com/threads/capacitor-question.41026/post-1061837

Here is a simple calculator- you can ignore the "L" for inductance, lets assume you want to filter to 80Hz on the input of your amp that has a 100K ohm input impedance

You would need a cap that is .019 mfd or nearest commercial value 0.022 is good enough for a first order filter

Reactance calculator R L C inductance inductor capacitance capacitor frequency - sengpielaudio Sengpiel Berlin

Reactance calculator R L C inductance inductor capacitance capacitor frequencyr - Eberhard Sengpiel sengpielaudio

sengpielaudio.com

I found a discussion by the late Roger Modjeski on this exact topic that dispels the need for accurate 1% caps- btw Roger had no time for exotic capacitor material talk fwiw


link:

passive crossovers

passive crossovers

www.audiocircle.com

"A quick note on how to calculate crossover component values for 6 dB/octave filters.

C=1/6.3 x frequency x input impedance of amplifier (or thing following the filter). It's simple.

A customer requested a 600 Hz crossover at the input of an EM7 amp. Standard EM7 amps have 150K input impedance so the value would be 1.76 nanofarad (1760 pf). The closest standard value is .0018 uF. Anything from .0015 to .0022 would be fine. For gradual crossovers the value is not critical. The only place accurate (1%) capacitors are needed is in a RIAA network or an active filter. I find it amusing when neuoritoc audiophiles want to use a 1% capacitor where the value is approximate anyway. That's just plain silly.

(more on this if you want)

He also asked about the RM-10 which has 100K input impedance. Most power amps are in the range from 15K to 150K ohms and you can usually check this by simply plugging an ohm meter into the input jack of the amplifier in question. If you get on OL on the meter or several meg ohms there is probably a capacitor in the input already and you have to get to the other side of that cap, measure the resistance and change that cap to the new value. That's all there is to it.

Any capacitor of any voltage will work fine. I would use a polypropylene or Mylar."

Good luck!

 

[Ron Resnick]

Ron this is quite easy and your are correct it needn't be over complex.

I tried to articulate it here- your are applying a simple 1st order filter to the input of the amp:

Capacitor Question

I tried to custom order 2,700pF and 3,000pF capacitors from Duelund, from its CAST copper line, for a DIY high-pass filter. Duelund says it cannot guarantee each capacitor any closer than 20% to the actual value requested. In theory I could have two capacitors that are 40% apart in value! I...

www.whatsbestforum.com

This is not an active filter like @DasguteOhr suggests here:

https://www.whatsbestforum.com/threads/capacitor-question.41026/post-1061837

Here is a simple calculator- you can ignore the "L" for inductance, lets assume you want to filter to 80Hz on the input of your amp that has a 100K ohm input impedance

You would need a cap that is .019 mfd or nearest commercial value 0.022 is good enough for a first order filter

Reactance calculator R L C inductance inductor capacitance capacitor frequency - sengpielaudio Sengpiel Berlin

Reactance calculator R L C inductance inductor capacitance capacitor frequencyr - Eberhard Sengpiel sengpielaudio

sengpielaudio.com

I found a discussion by the late Roger Modjeski on this exact topic that dispels the need for accurate 1% caps- btw Roger had no time for exotic capacitor material talk fwiw


link:

passive crossovers

passive crossovers

www.audiocircle.com

"A quick note on how to calculate crossover component values for 6 dB/octave filters.

C=1/6.3 x frequency x input impedance of amplifier (or thing following the filter). It's simple.

A customer requested a 600 Hz crossover at the input of an EM7 amp. Standard EM7 amps have 150K input impedance so the value would be 1.76 nanofarad (1760 pf). The closest standard value is .0018 uF. Anything from .0015 to .0022 would be fine. For gradual crossovers the value is not critical. The only place accurate (1%) capacitors are needed is in a RIAA network or an active filter. I find it amusing when neuoritoc audiophiles want to use a 1% capacitor where the value is approximate anyway. That's just plain silly.

(more on this if you want)

He also asked about the RM-10 which has 100K input impedance. Most power amps are in the range from 15K to 150K ohms and you can usually check this by simply plugging an ohm meter into the input jack of the amplifier in question. If you get on OL on the meter or several meg ohms there is probably a capacitor in the input already and you have to get to the other side of that cap, measure the resistance and change that cap to the new value. That's all there is to it.

Any capacitor of any voltage will work fine. I would use a polypropylene or Mylar."

Good luck!

Thank you very much!

I'm not worried about one percent matching. But Duelund's tolerance of plus or minus 20% from requested value seemed Mickey Mouse.

 

[VinylSavor]

You still have to measure the existing resistor if you wire another in parallel. This was discussed in a parallel thread. As you say, and easily confirmed reading the measurements section in Stereophile, most manufacturers only give an approximate value of input impedance.

Wiring a calculated much higher resistor in parallel to decrease the resistance value can be a possibility if the capacitor is chosen to have the closer higher value - 1% resistors are trivial, 1% capacitors are hard to get. But meters are needed, there are no subjective ways of doing it!

RxC must keep the same value, if C is larger we need a smaller R. Adding a parallel resistor decreases R.

Yes of course! The easy external solution: Measure the input resistance of the amp. Then equalise them by adding parallel resistances (can be in the external filter) and to match the capacitance to get the desired result. This way the tolerance between caps can be equalised as well!

The elegant and 'audiophile' solution was suggested as well: If there is already an input coupling cap in the amp, simply size this along with the following resistor to get the desired results. Easily done by a qualified technician and this way not a single additional component in the signal path. Easily reversible as well

 

[Holmz]

I tried to custom order 2,700pF and 3,000pF capacitors from Duelund, from its CAST copper line, for a DIY high-pass filter.

Duelund says it cannot guarantee each capacitor any closer than 20% to the actual value requested. In theory I could have two capacitors that are 40% apart in value! I think this is ridiculous.

Who makes a high quality high-end audio capacitor with values of 2,700pF and 3,000pF and some semblance of consistency?

Thank you.

Single ended or XLR
XLR: https://www.vandersteen.com/products/m7-hp

There is an RCA variety, and it is somewhere in my study.

If your runs are long, then you will want these out near the mono blocks.

 

[christensenleif@msn.com]

Then you have Marchand XM-46, very transparent
Audiophile Bill took up on my recommendation and use them to his Pnoe against his NNNN Devor 23 subhorn

XM46 Passive Line Level Crossover, PLLXO

Easy to install toggle switch on inputs for bypass function

 

[Alrainbow]

Ron have you gone on diy forums ??
There maybe people who own your products and can sell you matched specs after measuring them

 

[Ron Resnick]

Ron have you gone on diy forums ??
There maybe people who own your products and can sell you matched specs after measuring them

Thanks but I can buy from Jeff at Sonic Craft, a multi-decade professional who sells discrete components and does his own careful matching.

 

[Ron Resnick]

Single ended or XLR
XLR: https://www.vandersteen.com/products/m7-hp

There is an RCA variety, and it is somewhere in my study.

If your runs are long, then you will want these out near the mono blocks.

Then you have Marchand XM-46, very transparent
Audiophile Bill took up on my recommendation and use them to his Pnoe against his NNNN Devor 23 subhorn

XM46 Passive Line Level Crossover, PLLXO

Easy to install toggle switch on inputs for bypass function

I appreciate people trying to be helpful.

But I wondered how long it would take people to wander off the reservation into topics and suggestions which have only the slightest tangential relevance to the extremely specific question in the opening post.

 

[sbo6]

In a simple first order high pass filter my understanding is that only a capacitor is needed because the input impedance of the amplifier functions as the resistive load.


You are totally correct about this! But you are minimizing audiophilia nervosa: I want any additional components to be fancy, high-quality components.

Ron, I've been down the road I think you're headed down. A few things worth noting if you haven't yet - 1) measure the amp(s) input impedance, don't go solely by what the manufacturer states as you likely won't be able to return the caps. 2) First, purchase more modest priced caps as you might not get the integration you expect via your DIY first - pass. If it works, great, then buy the uber - expensive high - quality caps replacing the inexpensive ones.


Also, instead of the in - line cap, you could try this which is what I'm about to do: https://www.wilsonaudio.com/products/activ/activxo

 

[sbo6]

I appreciate people trying to be helpful.

But I wondered how long it would take people to wander off the reservation into topics and suggestions which have only the slightest tangential relevance to the extremely specific question in the opening post.

It's called human nature and conversation, makes for good rapport. Also, this is not a work environment where time = $. What exactly is the downside? Food for thought.