• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Townsend Isolda cable

Status
Not open for further replies.

Max Townshend

Active Member
Audio Company
Joined
Mar 22, 2019
Messages
129
Likes
30
Here is an image of the signal generator, set white noise
Signal In.jpg
 

Max Townshend

Active Member
Audio Company
Joined
Mar 22, 2019
Messages
129
Likes
30
No, you have tried to sell it. There's a difference.
In Internet slang, a troll is a person who starts flame wars or intentionally upsets people on the Internet by posting inflammatory and digressive,[1] extraneous, or off-topic messages in an online community (such as a newsgroup, forum, chat room, or blog) with the intent of provoking readers into displaying emotional responses[2] and normalizing tangential discussion,[3] either for the troll's amusement, or to achieve a specific result such as disrupting a rival's online activities or manipulating a political process.
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,936
Likes
3,521
I have measured the voltages in these two images. Input 55.5 mV and set the y axis to volts. Driven from the monitor speaker output of the 2i2 set to maximum.View attachment 95762
So on the 'flat strip apart' cable we measure 1.1mV @ 10kHz over the cable with a 55.5mV input voltage. That would be a cable loss of 0,17dB @ 10kHz, using a cable specifically crafted to increase cable losses? Do I have the numbers right?
 

Max Townshend

Active Member
Audio Company
Joined
Mar 22, 2019
Messages
129
Likes
30
55mv between red and black into dummy load. Black to black, 70uV at 20Hz, 1000uV at 10k = 23dB. This is similar as shown in fig 3.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I have measured the voltages in these two images. Input 55.5 mV and set the y axis to volts. Driven from the monitor speaker output of the 2i2 set to maximum.View attachment 95762View attachment 95763View attachment 95763

We cannot conclude anything from the posted plots. Why 2 of the same plots ?

A: We don't know if the dummy load or a resistor was used as a load.
B: mr. Townshend should have made a reference measurement where the input of the ADC should be connected to the output of the amp so we get a real reference in the form of the sweep. The 55.5mV says nothing, we don't know how sensitive the ADC was set, we don't know if the voltage level is referenced.
C: mr Townshend conveniently forgot to mention the cable length as well. (this could have given a clue about calibration)
D: About the dummy load: The posted schematic won't have the impedance posted alongside. Amir's sim is better. Why is the plot from mr Townshend incorrect or the schematic is incorrect (has a Boucherot filter installed ?)
E: Why not a comparo between regular speaker cable and the Isolda instead of the single strip of isolda cable which measures FAR worse that any speaker cable.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
In Internet slang, a troll is a person who starts flame wars or intentionally upsets people on the Internet by posting inflammatory and digressive,[1] extraneous, or off-topic messages in an online community (such as a newsgroup, forum, chat room, or blog) with the intent of provoking readers into displaying emotional responses[2] and normalizing tangential discussion,[3] either for the troll's amusement, or to achieve a specific result such as disrupting a rival's online activities or manipulating a political process.
Now do “scammer.”
 

Speedskater

Major Contributor
Joined
Mar 5, 2016
Messages
1,610
Likes
1,333
Location
Cleveland, Ohio USA
Again very poor test protocol.
a better plan:
a] attach 'cable under test' from the amp to the speaker or dummy load.
b] with the Fluke measuring across the amps terminals. using a 1000 Hz tone, set the amps output level at 1.90 V RMS.
c] measure the voltage at the speaker's terminals.
d] run a frequency response curve at the amp's terminals.
e] without changing any settings, run a frequency response curve at the speaker's terminals.
f] repeat with another cable.
 

Max Townshend

Active Member
Audio Company
Joined
Mar 22, 2019
Messages
129
Likes
30
We cannot conclude anything from the posted plots. Why 2 of the same plots ?

A: We don't know if the dummy load or a resistor was used as a load.
B: mr. Townshend should have made a reference measurement where the input of the ADC should be connected to the output of the amp so we get a real reference in the form of the sweep. The 55.5mV says nothing, we don't know how sensitive the ADC was set, we don't know if the voltage level is referenced.
C: mr Townshend conveniently forgot to mention the cable length as well. (this could have given a clue about calibration)
D: About the dummy load: The posted schematic won't have the impedance posted alongside. Amir's sim is better. Why is the plot from mr Townshend incorrect or the schematic is incorrect (has a Boucherot filter installed ?)
The conditions are as described in the paper.
I measured the voltage into the cables with the Fluke 179. It may be +/- 10%
We cannot conclude anything from the posted plots. Why 2 of the same plots ?

A: We don't know if the dummy load or a resistor was used as a load.
B: mr. Townshend should have made a reference measurement where the input of the ADC should be connected to the output of the amp so we get a real reference in the form of the sweep. The 55.5mV says nothing, we don't know how sensitive the ADC was set, we don't know if the voltage level is referenced.
C: mr Townshend conveniently forgot to mention the cable length as well. (this could have given a clue about calibration)
D: About the dummy load: The posted schematic won't have the impedance posted alongside. Amir's sim is better. Why is the plot from mr Townshend incorrect or the schematic is incorrect (has a Boucherot filter installed ?)
E: Why not a comparo between regular speaker cable and the Isolda instead of the single strip of isolda cable which measures FAR worse that any speaker cable.

If you want all those answers, do the experiment yourself and get your own answers.
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,936
Likes
3,521
55mv between red and black into dummy load. Black to black, 70uV at 20Hz, 1000uV at 10k = 23dB. This is similar as shown in fig 3.
Where do you see "70uV at 20Hz"? And how is 70uV cable loss 23dB of a 55mV input voltage?
2 flat strips apart marked.jpg
Update/Correction, didn't notice you mentioned 20Hz and not 20kHz (I was focused on 10kHz and higher where you have the highest loss). But the question remains; "how is 70uV cable loss 23dB of a 55mV input voltage"?
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
AC voltage on the Fluke 179 is specified only up to 1 kHz. It is not a suitable instrument for this measurement.

For the lower voltages the datasheet even states: 1.0 % + 3 (45 Hz to 500 Hz)
mr Townshend should have made a reference by measuring the input voltage of the CUT directly on the input with his 2i2.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The conditions are as described in the paper.
I measured the voltage into the cables with the Fluke 179. It may be +/- 10%
If you want all those answers, do the experiment yourself and get your own answers.

Why not disclose what you did, mention lengths, use a clear reference ?
How can we repeat the measurements when I don't want to buy 7m Isolda cable and YOU provide the measurements.
Are you being vague on purpose ?
The 55.5mV is pointless as a reference when you use white noise upto 20kHz as the Fluke won't register correctly above 500Hz ...
Totally not scientific at all.
Why not do this with 'normal' speaker cable with the same DC resistance ?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Where do you see "70uV at 20Hz"? And how is 70uV cable loss 23dB of a 55mV input voltage?
View attachment 95785

The plot shows 70uV at 20Hz, the problem is that when this measured voltage is not referenced and the 55mV reference is not correct it says nothing.
Also the cable length is of importance.
When mr. Townshend doesn't provide us with references all math doesn't make sense.
 

JustJones

Major Contributor
Forum Donor
Joined
Mar 31, 2020
Messages
1,726
Likes
2,391
I'm not an engineer these other guys/gals are doing a good job of trying to get technical answers I am just wondering why the end of your cable is plain copper wire and cheap terminators?

1606317686633.png
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,936
Likes
3,521
The plot shows 70uV at 20Hz.
Didn't notice he mentioned 20Hz and not 20kHz. I corrected my post. Still interested to know how he ends up with a 23dB cable loss, because that huge.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Still interested to know how he ends up with a 23dB cable loss, because that huge.

Without a real reference we have to trust him... Given the track record of being vague and incomplete ... :(
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,936
Likes
3,521
Without a real reference we have to trust him... Given the track record of being vague and incomplete ... :(
We don't need the reference as we now have a graph showing voltages. Input voltage 55.5mV, and 1.1mV cable loss @ 10kHz. What's the formula that gives 23dB cable loss using these input parameters?

(There might be some error margin on the 55.5mV input voltage, but that error would need to be really huge to get a 23dB cable loss).
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
We don't need the reference as we now have a graph showing voltages. Input voltage 55.5mV, and 1.1mV cable loss @ 10kHz. What's the formula that gives 23dB cable loss using these input parameters?

(There might be some error margin on the 55.5mV input voltage, but that error would need to be really huge to get a 23dB cable loss).
23 dB is the ratio of 70 mV to 1 V, the approximate graph values at 20 Hz and 10 kHz.

As I've already said, the 55.5 mV figure is guaranteed to be wrong since it was obtained with an instrument incapable of making this measurement.
 
Status
Not open for further replies.
Top Bottom