How important is the cable between the SUT and phono pre?

So this AU-340 sat on the sidelines for a few months and I just hooked it up with a nice quality 18" shielded interconnect. Wow, sounds spectacular with my retipped Denon DL-103S that Steve at VAS did for me. Used an elliptical stylus on the original cantilever, don't even have the alignment spot on yet. Listening to Willie Nelson, never sounded so good!

BillWojo

Thanks, but as I’ve said a couple times now that wasn’t the issue. 

@jpjones3318 

Don't know if this will help, but resistance is a DC measurement in Real Numbers, while impedance is an AC measurement in Complex Numbers. Capacitance and inductance both have non-zero imaginary components, and analysis in this domain requires Complex Arithmetic and/or calculus on the Complex Plane.

This isn't particularly hard, but it does require some exposure to the ideas.

After reading this thread I am glad my Parasound JC3 Jr. phonostage obviates the need for a SUT with my  .4mV output cartridge.

As important as phono-cables.

However, transformers do not reflect secondary load C to the primary.   

@jpjones3318 

Transformers transform impedance and impedance includes capacitance. However, things are not as simple as just that. You have to also consider how a parallel capacitance interacts with the inductance of the transformer- and here we are talking about a resonance. This is also occurring on the primary side! Of course, I'm saying the same thing that Dave @intactaudio just did in different words.

If your EE friends tell you this isn't a thing, here's an EE that is telling you it is, just FWIW... but 'multiplied back to the source' is only correct in that there is a fractional factor involved, variable with frequency. This is really one reason I stay away from SUTs if I can help it 😁

As a general rule of thumb, the best advice is to use low capacitance cables. If you do this right, the electrical resonances involved will be so high that on the cartridge side it will be well outside the transformer's bandwidth. If loaded properly on the secondary side (such that there is no ringing) again you're off the hook. At any rate don't think for a minute that the loading on the secondary side (including capacitive effects) can't be measured on the primary side!

No, that's not the problem I was having.  Worked through it with a friend and see that I was not correct.  Regardless, something that actually explains it would be beneficial.  I like to enhance my understanding of things, not just parrot what someone says because they said it. 

Would love to see a cite with a detailed technical explanation.

You are looking too deep into it.  This is the simplest basic transformer theory and the simple ideal transformer model is all that is required.  The problem you are having is assuming the load is a resistance but in reality it is an impedance which can be any combination of R, L or C.  By the time you get into the models that include external capacitive loads, the concern is more about how that capacitance interacts with the internal R,L and C of the transformer and not how it reflect back to the source.

dave

My laymans understanding of the scenario is that because the MC transformer converts high current low voltage LOMC to a higher voltage low current signal exiting the transformer an interconnect with high capacitance between the transformer and MM phono is more prone to noise intrusion.

@jpjones3318 

I think Chapter 1.3.2 or page 11 is what you looking for?

Thank you for the reference.  Section 1.2.4 is in regard to inter- and intra-winding capacitance and electro-static shielding, which I understand.  While Figure 16 depicts secondary load C as CL, I do not see a reference to it in the text.  This section doesn’t seem to address reflected C.  

For transformer winding capacitance, please read the article on Jensen Transformers Application Notes "AN008" - Audio Transformer by Bill Whitlock

https://www.jensen-transformers.com/application-notes/

Would love to see a cite with a detailed technical explanation.  I can’t find one, equivalent models I can find don’t show it, and a couple EEs I’ve talked to say it doesn’t.  Repeating a formula is not an explanation.  

However, transformers do not reflect secondary load C to the primary.   

They absolutely do reflect any capacitance seen on the secondary by the turns ratio squared.  Say you have 150pf (50p cable + 100p input)  at the secondary of a 1:20 SUT.  This will appear as 150*20*20 or 60,000pF or 60nF or 0.06µF to the source.

dave

Hi @imhififan, he's basically explaining a zobel like I mention above, and some generic advice that may or may not fit a circumstance.   However, transformers do not reflect secondary load C to the primary.   

MC step-up transformers explained

"The capacitance of the cable connecting an mc step-up transformer to the following phonostage also plays a part, which is why the interconnecting cable should be a low capacitance design and kept as short as is practical."

Impedance != capacitance.  

looking at transformer equivalent circuits I don't see reflected capacitance.  Reflect impedance, reflected leakage inductance, but not capacitance.  Do you have cite for this?

@jpjones3318 Transformers are called that because they transform impedance. They do not isolate it!

So a load on the output of the transformer very much affects the source on the input side. This is part of why the transformer must be loaded correctly. Incorrect loading of the transformer can have other effects too- for example all transformers have inter-winding capacitance. If the transformer is too 'lightly loaded' (output load impedance is way too high) the inter-winding capacitance will begin to affect frequency response, perhaps manifesting as a hump in the frequency response rather than being flat.

Dear @billwojo  : You need not only the short cable you can use but the best you own.

Now, I  still own the 340 and is a good SUT if and only if you make some changes on it: first are the input cheap RCA connectors for a decent ones, second that that short and high quality output IC cable be hard wired directly to the inside 340 board and third that all those internal input wires/cables connected to the 340 board and to the input connectors be changed by better quality wires.

Regards and enjoy the MUSIC NOT DISTORTIONS,

R.

Those changes makes/converts a good SUT in an excellent one that competes with any today top SUT.

Regards and enjoy the MUSIC NOT DISTORTIONS,

R.

To minimize capacitance, in addition to other measures, experiment with unshielded cable.  In some home environments, you can get away with it.

Try use max 0.5m length.

im using Belden 8402 and easy to find on eBay .

better than kimber tak phono .

Enjoy 

Agree with @mijostyn keeping your run out of the SUT super short (0.5m or less) is probably what you should focus on in most cases. 

@jpjones3318 No I don't have a cite for that part. I've seen it mentioned a couple times on forums, but forget where / who. I took that, combined with the info on the Hagerman page (which was fascinating to me), and my own SUT + cable listening tests (more capacitance, longer runs after SUT == BAD) and ran with it haha. 

@billwojo , The single most important factor is to keep the cable as short as possible. I would use a high bandwidth balanced cable like Canare Da206. 

I use Sowter transformers which I mounted inside the phono stage with a switch to go back and forth between MM and MC. Doing this depends on how much real estate is available inside the phono stage. The transformers are very small. 

https://www.sowter.co.uk/acatalog/E_A_Sowter_Ltd_PHONO_CARTRIDGE_TRANSFORMERS_30.html

@jasonbourne52, it's not quite that simple for MM as increasing capacitance will also move the resonant frequency down, depending on the electrical characteristics of the generator, of course.  This is why on a cartridge like the AT150MLX, if you want flat response, the cartridge cannot see more than 150pF.  Also why that cartridge has such a maligned reputation for exaggerated HF response.

@mulveling, looking at transformer equivalent circuits I don't see reflected capacitance.  Reflect impedance, reflected leakage inductance, but not capacitance.  Do you have cite for this?

@billwojo, capacitance on the secondary with load R will form a zobel, which can tune the response of the transformer.  For best performance every SUT should be tuned this way, but as it is cartridge, transformer, and upstream phono stage dependent, it's not practical for the average person.  Just know that changing C on the secondary can have an audible effect, but it's to do with the transformer response itself.  

At the bottom of this page is a tech note from Morgan Jones: https://www.sowter.co.uk/specs/8055.htm

@billwojo

I am probably not the kind of expert you need here. But look at this page which I found very interesting:

http://www.hagtech.com/loading.html

The first section deals specifically with loading MM cartridges with capacitance. However, the formula Fr = 1 / (2*PI*SQRT(L * C)) works equally for MC and MM cartridges. This formula defines the cart’s resonant frequency based on its inherent inductance and the downstream (cable) capacitance. The resonance is often accompanied by a sharp peak and followed by rolloff. Generally you want this frequency as high as possible, well out of the audible band (> 20kHz). This generally requires low capacitance OR low inductance (as per the formula). MC cartridges get the low inductance (that’s the L in the formula) for "free". Normally this is enough that they need not concern themselves with phono cable capacitance. HOWEVER with a SUT reflecting the output cable’s capacitance times the square of its turns ratio (look this up elsewhere), capacitance could suddenly become large enough to drop the resonant frequency back near or into audible range and become a concern. For example: say you (unwisely) choose a fat audiophile interconnect run with 200 pF capacitance. And your SUT is a 30x. 30 * 30 * 200pF = 180,000 pF. Wow. Say your MC has an inductance of 1mH. Using the calculator in that Hagerman page I linked, the resonance is now at 3,800 Hz. You’re gonna have a bad time! Choose a cable with 40pF and now it’s at 26,000 Hz. Much better!

You have multiple cables. In the time spent asking for guesses here you could hook them up and know. Then let us know. Why not try them all, let us know what you tried, how it sounded?

@mulveling

"MC are said to not be sensitive to capacitance, but the problem with a SUT is that it reflects the capacitance of its output cable times the SQUARE of its turns ratio. That’s enough capacitance that it WILL start to affect the MC cartridge."

Can you point me to some more information on this? I was searching for information and found nothing to indicate that this was a problem. I've read that the cables from cartridge to SUT didn't need to be particularly low capacitance but nobody talks about the cables from SUT to phono pre.

Thanks

BillWojo
 

Really, really important. Ultra low capacitance, and low run length < 1.0m are key. You need to try to keep it under 50 pF. Don’t use a "normal" thick audiophile interconnect.

Before I knew how important low capacitance was, I did some cable rolling with high quality "normal" interconnects. All the 1m cables had their sonic signature GREATLY magnified, to the point of obvious coloration. The AQ Fire 1m became overly detailed and aggressive. The AQ Sky 1m sounded overly warm, thick and and bloomy. The 0.5m AQ Niagara, normally inferior to the higher-end Fire and Sky, did better and was relatively uncolored - because of the halved length and significantly lower capacitance. The the dedicated AQ WEL LP phono cable (optimized for lower capacitance), 0.5m, blew them all away. The Bob’s Devices 0.7m interconnects are also really good (but the WEL LP is still better) - I have both their copper and silver cables. They’re optimized for SUT use, obviously.

This was all tested on the same Koetsu cartridge and Bob’s Devices Sky SUT.

MC are said to not be sensitive to capacitance, but the problem with a SUT is that it reflects the capacitance of its output cable times the SQUARE of its turns ratio. That’s enough capacitance that it WILL start to affect the MC cartridge.

Many Cardas interconnects have very, very low capacitance - so if you've got those in your collection, give 'em a try. I believe Bob's uses Cardas ingredients to make his cables.

It has a bypass switch but right now I'm just using various versions of the Denon DL 103 cartridges. So 

MC all the way. I have plenty of good shielded cables so I should have what I need. 

Thanks

BillWojo

If the capacitive load is below the minimum for a particular mm cartridge it will have a high frequency peak. Moving coil cartridges are immune to capacitive loading.

Some mm cartridges have a recommended capacitive load for flattest response. This capacitance is the sum of the tone arm wiring, cable to the phono stage and the phono stage input capacitances.

Does the Denon AU-340 have a bypass switch so you can run a mm cartridge through it and avoid the transformer? Cable capacitance will only affect the high frequencies of mm cartridges by rolling them off.

You want well shield and low capacitance.

But at the end of the day just try your best cables - you ears will tell you what works. You want no buzz hum or noise.