You are here

Big Knob passive distortion.

Customising, building or repairing your own gear? Need help with acoustic treatment or soundproofing? Ask away…

Big Knob passive distortion.

Postby ef37a » Sun Aug 30, 2020 10:01 am

Impossible right? 'Sok, have not lost my marbles (yet)

I answered a thread about this at HR.com. A well respected guy there found when driving the BKp hard from a Focusrite 8i6 the sound would get distinctly 'harder' and pushing even more, definite distortion. He is not a snot[-nosed newb that will have some ill conceived rig and doing a silly thing.

I have a BKp (and an old 8i6*) so I decided to investigate. NB my NI KA6 drives it and I have never had a problem through my Tannoy 5As. I metered the TRS inputs and tip to sleeve was O/C (goes via switch to caps then VC) but! The ring read just 1K Ohms on my Fluke 83.

That means the 'ring' side of the 8i6 balanced output is loaded well below the accepted 10k, why? If the AI uses a version of the quite acceptable TL072 op amp this is where the distortion arises since they really don't like driving a load much lower than 2k and preferably much higher. The reason I get no problems with the KA6 is because the outputs are impedance balanced.

Would be interested in comments, especially Hugh's.

*I shall hook this up in the next day or so. I had it working weeks ago then decided to play with MixControl! Got lost, got mad and have not touched it since. I shall gird loins and have another bash.

Dave.
ef37a
Jedi Poster
Posts: 12036
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Big Knob passive distortion.

Postby Hugh Robjohns » Sun Aug 30, 2020 11:49 am

ef37a wrote:Impossible right?

No, not at all. If the load impedance becomes too low for the driving device -- especially when putting out a very high signal voltage -- the driver could very easily struggle and generate audible distortion.

There's also a high possibility of the BKp not presenting a properly balanced load and that might also cause problems.

It all depends on how the BKp is engineered internally, but not many (any?) modern passive volume controls / monitor controllers are engineered to present a constant-impedance, correctly balanced load. The circuitry, such as it is, will be a much less expensive 'approximation'.

Passive controllers can be useful and are cheap... but their performance is inherently unpredictable and compromised in some situations.
User avatar
Hugh Robjohns
Moderator
Posts: 28361
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Big Knob passive distortion.

Postby ef37a » Sun Aug 30, 2020 12:02 pm

Thank you Hugh but I meant that it is impossible for a passive device OF ITSELF to generate distortion (and I include diodes and VDRs as at least not "passive" devices!)

I was more wondering if you had any theories as to why they have put a 1k load on the ring/pin 3 circuit? The device is not balanced (outs are impedance balanced but of course that is imperfect relying as it does on the source and the position of the pot) so why not simply leave the ring circuit O/C?

I bought mine mainly for the input switching facility and it sits barely a foot from the KA6 so cables are very short.

Dave.
ef37a
Jedi Poster
Posts: 12036
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Big Knob passive distortion.

Postby Hugh Robjohns » Sun Aug 30, 2020 12:30 pm

ef37a wrote:Thank you Hugh but I meant that it is impossible for a passive device OF ITSELF to generate distortion...

Well yes... but as an isolated device it's not a lot of use to anyone! It's the complete signal chain that is important, and a passive attenuator can cause problems in some situations, as evidenced by your correspondent.

I was more wondering if you had any theories as to why they have put a 1k load on the ring/pin 3 circuit?

No idea. I'd need to take one apart and figure out the full schematic and then have a think about what's going on in there...

You seem to be implying that it is configured as a single-sided potential divider, and if that's the case I imagine the 1K on the cold side is just to tie down that side of the source's balanced output in case it's from a floating transformer output or a cross-linked active symmetrical stage.
User avatar
Hugh Robjohns
Moderator
Posts: 28361
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Big Knob passive distortion.

Postby James Perrett » Sun Aug 30, 2020 12:56 pm

I'm not quite clear...

Are you saying that there is 1k between the ring and sleeve connector or between the tip and ring connector? I would have thought that an audio interface with TRS outputs shouldn't have a problem with a short circuit between the ring and sleeve because there's a strong likelihood that someone is going to plug a TS jack into that output at some point.

It would be interesting to measure the impedance between the tip and ring - you could probably rig something up with a variable resistor in series with the ring input to give you an approximate value.
User avatar
James Perrett
Moderator
Posts: 9694
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Re: Big Knob passive distortion.

Postby ef37a » Sun Aug 30, 2020 1:14 pm

Hi James, I have put it back together now and I am not taking the 14 screws, 9 nuts and allen screw off again! Sorry!

The 1k is a physical resistor between ring and sleeve. The tip goes off to two 10uF capacitors so I would have to rig some sort of AC impedance tester to check that (spose I could do that externally and see what the input Z is at say 1kHz)

Yes, a TS plug into a 'proper' balanced output would short the amp driving the ring and we should not do it! I have actually refined my thinking about impedance balanced outs, certainly for the home 'projjy' market. They can easily supply enough drive voltage and have the advantage that the newb cannot get a polarity reversal and now they are proof against Big Knobs!

Dave.
ef37a
Jedi Poster
Posts: 12036
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Big Knob passive distortion.

Postby James Perrett » Sun Aug 30, 2020 1:43 pm

ef37a wrote:The 1k is a physical resistor between ring and sleeve.

Ah - yes. That's going to be a real test of your interface's output stage.

Even an impedance balanced output will still see the low impedance as the signal feeds the 1k plus the balancing resistor in your interface (150 ohms?). So your experience demonstrates that the KA-6 output stage is better at driving a low impedance load than the Focusrite.
User avatar
James Perrett
Moderator
Posts: 9694
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Re: Big Knob passive distortion.

Postby ef37a » Tue Sep 01, 2020 7:02 am

Some roughly obtained data on the BKp.

Insertion loss with the pot at max turns out to be 1.9dB into an open circuit but see further...

The input (tip) resistance is 10.53k so just the pot but that means the output resistance could be 2.5k or a bit more and so loss will depend upon load.

Cables to monitors should also be kept short, mine are under 2mtrs.

Dave.
ef37a
Jedi Poster
Posts: 12036
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Big Knob passive distortion.

Postby Folderol » Tue Sep 01, 2020 8:30 am

All in all, that sounds like a pretty cheap and nasty design. However I wonder if the 1k is in fact a mistake. With a mixture of 4 and 5 band resistors it is very easy to confuse 1k and 10k.
User avatar
Folderol
Jedi Poster
Posts: 11367
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Big Knob passive distortion.

Postby ef37a » Tue Sep 01, 2020 9:11 am

Folderol wrote:All in all, that sounds like a pretty cheap and nasty design. However I wonder if the 1k is in fact a mistake. With a mixture of 4 and 5 band resistors it is very easy to confuse 1k and 10k.

Or! even worse, people that cannot read the colour code at all. When I was a bench solderer at the network factory there were but two resistor values on the blue tubs back of benches. One contained 470k 0.25W drain resistor in a telephone 'master' module and the other 100R 0.25W for the terminating Rs in an ISDN module. To the lovely ladies they were both just "little blobs with wires on" !

Some re-work was needed but **'knows how many modules went out with the wrong resistors fitted before they had an electronics tech on the factory floor!

"Cheap and nasty" Will? In some ways yes but it is built like a BSH. The jacks are pretty good quality, not the 'bent tin' variety you get on US amps and guitars!

As I think I said, I bought it for the input switching, could have made something but when you factor in case, switches, 8 jacks.....

Dave.
ef37a
Jedi Poster
Posts: 12036
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Big Knob passive distortion.

Postby cyrano.mac » Wed Sep 02, 2020 4:04 pm

I'm guessing the 1 k in the Focusrite is a series resistor to protect one leg of the output amp from being short-circuited?
cyrano.mac
Regular
Posts: 130
Joined: Sun Jun 02, 2019 5:00 pm

Re: Big Knob passive distortion.

Postby ef37a » Wed Sep 02, 2020 7:26 pm

cyrano.mac wrote:I'm guessing the 1 k in the Focusrite is a series resistor to protect one leg of the output amp from being short-circuited?

No Sir. The 1k is on the INput jack and takes the ring to sleeve, i.e. ground and thus loads the ring driving amp. if there is one. Most op amps such as the NE5532 and LM4562 will not mind in the least but any of the TL0 series will and probably several others.

In any case "10k" is THE sort of standard lowest input impedance for line inputs. A few manfctrs are a bit haughty at 5kish but 1k is very naughty indeed IMHO!

Dave.
ef37a
Jedi Poster
Posts: 12036
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk