John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Jump to Latest
#76,227
Nothing as complicated as that. Some amps don't like seeing certain loads at RF frequencies; a design error in my view. However, given one of these amps you can sometimes rescue the situation by ensuring that the speaker cable looks vaguely resistive for RF. A simple way to do this is to add a CR at the speaker end, so the cable sees something like a resistance at RF frequencies. Ideally the R should roughly match the cable RF characteristic impedance - often around 50-100ohms for normal speaker cable but lower for some abnormal speaker cables.
 
#76,228
DF96 said:
Nothing as complicated as that. Some amps don't like seeing certain loads at RF frequencies; a design error in my view. However, given one of these amps you can sometimes rescue the situation by ensuring that the speaker cable looks vaguely resistive for RF. A simple way to do this is to add a CR at the speaker end, so the cable sees something like a resistance at RF frequencies. Ideally the R should roughly match the cable RF characteristic impedance - often around 50-100ohms for normal speaker cable but lower for some abnormal speaker cables.
Click to expand...
Yes.

Particular/SOT values of CR should be at both ends of the cable ?.
Amplifier build out L is another part of the equation.

Dan.
 
Last edited:
#76,231
DF96 said:
Nothing as complicated as that. Some amps don't like seeing certain loads at RF frequencies; a design error in my view. However, given one of these amps you can sometimes rescue the situation by ensuring that the speaker cable looks vaguely resistive for RF. A simple way to do this is to add a CR at the speaker end, so the cable sees something like a resistance at RF frequencies. Ideally the R should roughly match the cable RF characteristic impedance - often around 50-100ohms for normal speaker cable but lower for some abnormal speaker cables.
Click to expand...
R.Cordell has a short secton in his book. I can't recall if it came up in the interviews series.

I think the RC at the speaker end and across the amplifier's speaker terminals will also work for interference impinging on the speaker cable.

I have only added the amp terminals RC (the Pi version of the Thiele Network).
Maybe I should try the speaker end as well.
 
Last edited:
#76,232
DF96 said:
CR at speaker end will do. An 80ohm cable with 80ohms across the far end will look like 80ohms at the near end too.

Alternatively, use a proper Zobel network at the amplifier end so it doesn't see the cable at RF frequencies.
Click to expand...

AndrewT said:
R.Cordell has a short secton in his book. I can't recall if it came up in the interviews series.

I think the RC at the speaker end and across the amplifier's speaker terminals will also work for interference impinging on the speaker cable.

I have only added the amp terminals RC (the Pi version of the Thiele Network).
Maybe I should try the speaker end as well.
Click to expand...
IME putting "ideal" (large value ex military styro) CR netwoks across each of two drivers paid huge dividend in terms of perceived distortion, depth information and power handling....'downside' was source polarity dependence...ie programme AP is revealed.

Dan.
 
#76,237
nezbleu said:
If a cable makes the load more difficult it should be discarded.
Click to expand...
Except that at rf the primary determinant of interaction with the amplifier is simply the speaker cable's exact length, modified by other TL parameters and the load. So in the case of an rf unstable amplifier, it's a crapshoot as to whether any specific cable might or might not interact with any specific unstable amplifier. I'm with DF96, it's an amplifier design issue and that is where the solution should lie, so that it's unconditionally stable.
 
#76,238
Some expensive cables may be difficult to drive. An example might be one which foolishly tries to provide 8ohms characteristic impedance (not realistically possible at audio frequencies!). This would provide a low and capacitive reactance at RF, so could upset an amplifier more than a normal cable with a higher impedance.

So if an amp misbehaves with an ordinary cable it is the amp's fault. If it misbehaves with an exotic cable then it is the cable's fault.
 
#76,239
DF96 said:
If your amp is well-designed then there is no need. You already have a pi network at the amp end so that will partially isolate the cable for RF.
Click to expand...
Interesting that the speaker cables are not viewed as an antenna connected to the amp's output. It would be interesting (well at least to me) to know how many of these wire gurus have solid backrounds in rf or transmission line theory .:2c:
 
#76,240
In my exoerience, owners of reel to reel, and even the better cassette decks, were by and large not convinced by most cable vendors. Just as owners of setups which needed to run uncommonly long speaker cables (due to room architectural reasons) mostly stayed with normal cable, perhaps of a larger gauge (as in my case, 2*5.5 mm2, 2*256 Litz wires, but I am forced to use 2*6m or 2*19 ft cable runs, and I love them there amps being able to hop along).
 
Last edited:
Status
Not open for further replies.
Top Bottom