New pots old pots, anyone figured out what kind of capacitance a pot adds ?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Tubetec

Well-known member
Joined
Nov 18, 2015
Messages
5,951
Just wondering if anyone has studied into how much capacitance an old fashioned big metal pot  might have vs a modern plastic mini pot?
Obviously the value of the pot will have a major effect on any capacitance ,higher values mean more treble leaked to ground .
My question arises in relation to the Federal Am-864/u which has a dual 500k log pot at the input, of course if the unit was used as an Am limiter, audio above around 5khz would make little difference either way ,but I'm working on building a Federal with modern audio transformers ,so Im wondering if the dual 500k pot at the input will kill the highs, Ive seen one person who used a 2x50k pot,but Im not sure if this will effect the circuit operation .
Any insight into this would be great.
 
Depending on where the capacitance appears in the circuit, it is not even certain the treble will 'leak to ground' as you put it. If there is more capacitance between the top of the pot and the wiper than there is between the wiper and ground then you will actually get treble boost.

Cheers

Ian
 
Thanks Jr and Ian ,scuse the 'how long is a piece of string ' nature of my  question .I  try to avoid attenuation where ever possible , and generally try to design my gear with cm's between input socket and grid. A 500k pot hanging off a transformer secondary  doesnt sound like a recipe for wide bandwidth in any case ,at least at higher attenuation settings .So, in the case of the federal 864/u, 600 ohms to 10k transformer meeting a load of 1meg ,thats a factor of 100 times the impedence,seems overkill to me .
100k seems far more sensible as grid resistance ,most of the time Id imagine Id use a 10k :10k transformer ,certainly with line input and most likely with condenser mics/ higher level sources .Ive toyed with the idea of using a quad gang pot and switching in the various sections to provide a range of impedence matching options ,course more wipers , carbon tracks and extra metalwork just adds capacitance ,so law of diminishing returns kicks in . Im just trying to work out the best option before bullet holeing my quad II chassis to mount the pots .
I see you have some experience with the carnhills(Ian) ,what would be your opinion on using the carnhill vtb  2291 as an output transformer for the 864/u, driven by the 6sn7 push pull ? of course it will normally  only ever have to drive line level impedences.
Bridge rectified the 6.3volt winding of my quad ,added a .47 ohm in series with each phase ,that gets me 6,2 volts dc across the pair of 6sk7's ,6sn7 and 6sq7 heaters. I'll use a small reservoir cap on the ht and two chokes  to reduce the volts from the quad mains transformer  to around 250 for the output stage and rc down the voltage further for the input  and detector stages.
Found this on ebay also ,
http://www.ebay.ie/itm/Classic-Circuit-Tube-Preamplifier-Preamp-Board-DIY-Kits-For-12AX7-21AU7
wonder what the origin of this circuit is ?
would it work if I used both sides of this pre in balanced push pull with transformer coupleing?
Time to get to bed for now ,but thanks for your reply's.


 
I just took a look at the federal 864/u schematic. Like most vari-mu limiters, the control voltage is applied to the centre tap of the input transformer. Normally the secondaries of the transformer would be connected direct to the vari-mu tube grids. However, in this design, a dual 500K pot is used as an input level attenuator. If the transformer is a 600:10K, then each 500K pot sits across half the secondary  so it looks like a 2K5 source. Compared to 500K this is nearly zero so the maximum output resistance of the pot will be a quarter of this or 125K. Even if there was 40pF hanging off this, the response would only be 3dB down above 30KHz

I see no reason in principle why these pots should not be reduced to 50K

Cheers

Ian
 
Tubetec said:
500k pot hanging off a transformer secondary  doesnt sound like a recipe for wide bandwidth in any case ,at least at higher attenuation settings .
As Ian hinted at, the worst happens at 6dB attenuation, where the equivalent resistance is highest.
 
Electric guitars use high value pots, usually 250k or 500k, and the pots themselves don't tend to affect tone. The cable to the amp often does, as it has substantial capacitance (in the hundreds of pF). Turning down the volume from the full on position cuts the treble because the signal goes through a higher resistance (the increased resistance of the pot) driving the cable capacitance.

I can't imagine a pot by itself adds enough capacitance to worry about for audio.
 
Thanks for the replys gents,
Hmmm 30khz bandwidth , a bit below what good transformers can achieve nowadays, Im used to getting at least 80-100khz on simple single ended tube pre's with  high quality output transformer
I think the situation with regards guitar pickups might be a little different ,the coils measure maybe 5k for a single coil upto about 15k ohms for a hotwired humbucker, there is a slight difference in the tone between say a 250k pot and a 500k and maybe a few db's level difference also ,but yes your right any capacitance in the pot itself is going to be small compared to the 50pf per meter that you would find in a single core screened lead .
The Federal/Quad project is coming along nicely  now though , it should have a lot less wireing than the original ,so hopefully less incidental capacitance ,less hum induced in the wireing and more bandwidth.
Is there anybody who keeps the edcor range of Tx's in stock in the Uk ,or do people make their orders direct from the Us? the price is very competitive ,and many give good reports on them it seems .Im a big sowter fan ,but there pricey.
 
Tubetec said:
I think the situation with regards guitar pickups might be a little different ,the coils measure maybe 5k for a single coil upto about 15k ohms for a hotwired humbucker,
DC resistance is not the dominant factor, it's the inductance of several Henries. A typical single coil has about 2-3 H inductance; at 10kHz, that's a reactance of 120-180 kohm.


there is a slight difference in the tone between say a 250k pot and a 500k and maybe a few db's level difference also ,but yes your right any capacitance in the pot itself is going to be small compared to the 50pf per meter that you would find in a single core screened lead .
That's why you won't see a length of shielded cable between the transformer secondary and the grid in pro audio tube equipment. Wiring is done at the shortest.
Typical cable capacitance is closer to 50pF per foot than per meter.
 
Tubetec said:
Thanks for the replys gents,
Hmmm 30khz bandwidth , a bit below what good transformers can achieve nowadays, Im used to getting at least 80-100khz on simple single ended tube pre's with  high quality output transformer

Line output transformers are relatively easy. It is the input transformer where the HF tends to be limited, Good quality mic transformers for tubes (1:10 ratio) will reach 40KHZ but not 80KHz.

Cheers

Ian
 

Latest posts

Back
Top