Page 1 of 1

SUT Selection

Posted: Sat Nov 22, 2008 12:03 am
by dave slagle
There seems to be a lot of confusion on how to select a transformer for a particular MC cartridge. Many people consider this black art and I have to admit I have been shocked by the sonic differences from devices with similar electrical characteristics. I'll try to be brief and cover some of the relationships in a few short paragraphs.

It's about the turns ratio.

The first thing you need to know about the SUT in question is the turns ratio. This is an easy number to find out and is the most important. The idea here is the turns ratio tells you the gain but it also plays a major part in the load the cartridge sees. The rule is impedance is the square of the turns ratio. If you have a 1:10 turns ratio, you get a 1:100 impedance ratio. If you assume that the 1:10 will feed a phono stage with a 47K input resistor the 47K will reflect back 47K divided by the impedance ratio (100) or 470 ohms. It is really that simple. Many transformers (particularly the vintage ones) are specified with impedance numbers like 150:50K which simply translates to a 1:18 step up ratio. (sqrt(50000/150)). The impedance numbers are somewhat important since they suggest the ballpark of where the unit was designed to operate, but it is easy enough to toss that info out the door and measure things for yourself. One other important thing to realize about the turns ratio is that "More" isn't always better. It is surprisingly easy to have too much of a step up resulting in a situation where you overdrive (clip) your phono stage.

Relating the turns ratio to load.

Sticking with the above example of a 1:10 driving the 47K input, our cartridge in a perfect world sees a load of 470 ohms. Now what if we have a 103R with a 15 ohm internal impedance and we want to play with larger loads. (A larger load is one that is smaller in value) The traditional way to do this is to add additional resistance in parallel with the 47K to get the desired value. Again this is a simple application of the turns ratio. If we desire a 150 ohm load we would need to parallel a 22K resistor with the existing 47K, which nets us 15K across the secondary. (47000||22000=15000) Dividing the 15K by our impedance ratio of 100 nets us our desired 150 ohm load.

Houston, We have a problem

Everything here is nice and clean upon first glance, but we have actually hit our first speed bump. When we terminate the transformer with a different value, we not only change the load seen by the cartridge we change the behavior of the transformer itself! This means we are changing two parameters which creates a very unpredictable situation which goes a long way to explain why results of playing with secondary loading on SUT's has lead to such varied results since you cannot be sure what you are fixing.

Why the need for a load anyways.

The simple answer is all cartridges have a peak (resonance) at some high frequency and by increasing the load the peak is damped smoothing out the measured response. The general effect is when the load it too low (say unloaded) the cartridge sounds bright and when you take it to the other extreme and increase the load the highs start to sound rolled of. At some point in the middle the "sweet spot" is found and the only way I know how to this is in system by ear. The 800 pound gorilla sitting in the corner is the fact that the SUT will show the exact same behavior as the cartridge and loading will have a similar effect. Typically the resonance in the SUT will be an octave to a decade higher than that of the cartridge. Unfortunately there is no way to know if your choice of secondary load has damped the resonance of the cartridge taming the highs or simply rolled of the SUT masking the brightness of the cartridge resonance.

Gorilla Wrangling

How do we tame that gorilla in the corner? Menno van der Veen presented the best approach I have seen in his white paper on his SUT’s. Essentially his approach is to determine the load needed to make the transformer behave as desired and then add any additional load required by the cartridge to the primary of the transformer. He even goes as far as to measure his SUT’s under a number of conditions and provides the needed loading info for various impedance cartridges. This is the little brother of the 800 pound gorilla, lets call him the 400 pound gorilla on the couch. Just as increasing the load on the secondary damps transformer resonances (ringing) decreasing the source (cartridge) impedance tends to increase ringing. This makes the once simple loading of a transformer become a far more complex relationship. A few other gorillas strolling about the room involve the belief by some that the loading of transformers can cause more sonic harm than the problem it fixes and the idea that loading a transformer secondary causes phase shift at high frequencies.

Jane Goodall to the rescue.

Rather than ignore all of these primates strolling around our listening rooms the best approach is to attempt to understand them and learn how to live with them. The best way I know how to do this is to converse about experiences and understand that ultimately it comes down to first understanding ones musical preferences so their experiences can be related to the info available to us.

A final anecdote.

For some time now I have been a fan of primary loading of the SUT. Over the past few years I have encourages a number of people to play with primary vs. secondary loading and the results have been mixed. The one interesting thing was that consistently the people who preferred secondary loading said primary loading was “harsh� sounding and the people who preferred primary loading stated it brought more of the music out of their system. In looking at the measured response of the transformers in question with the known source and load impedances it quickly became apparent that the people needing the secondary loads were not only damping the ringing in their cartridges, but also taming a peak in response in the transformer. By loading the primary they actually made the transformer ringing worse, which is a very plausible explanation of what they heard. When looking at the measured response of the transformers used by those who preferred primary loading, the resonant peak was reasonably controlled and substantially beyond the audio band.

dave

Posted: Thu Nov 27, 2008 6:48 pm
by TonyB
Excellent write up, Dave!

That could explain why I never got my Jensen JT-346 MC step up to sound how I would like. The manufacturer recommends a secondary load of 6k81 Ohm in parallel with 4k12 Ohm + 680pF. I always thought that 6k81 Ohms is a too heavy load and I did not want to put an RC circuit on the secondary. Maybe I should and finish the cartridge loading on the primary.

I just wonder whether both the R in parallel with RC are the required transformer load or whether the 6k81 Ohms is the cartridge load.

Regards,

TonyB

Posted: Fri Nov 28, 2008 4:15 am
by dave slagle
Hey again :-)

I am not a big fan of applying any load to any transformer since it always seems to wreck the sound for me. Obviously certain transformers need to be designed with a load in mind (outputs feeding a speaker) but beyond that secondary loading is simply a band-aid to fix a problem that shouldn't exist or possibly doesn't matter.

If you give me a random transformer, I can run the tests and determine what conditions it will behave under. It is this type of revere engineering that dominates the audio world. My thought is tell me the situation you plan on using the transformer and within reason I can come pretty close to designing a device that will cut (round) the appropriate corners.

dave

Posted: Thu Feb 19, 2009 4:05 pm
by Qman
This is my first post to the forum. I'm using the 1:18(150ohm) primary of the Cinemags because it is a better match gain wise with my system. I have two cartridges with coil impedance's of 3.5ohm & 8ohm respectively and I would like to experiment using Vendereen's method of loading the primary side down to 20ohms or so. Can I just plug the 150 value into his formula?

His example based on his SUT's primary of 344ohm loading to 150ohm.

Rin=(344*150)/(344-150).
Rin= 265ohm.

Cinemag:
Rin=(150*20)/(150-20).
Rin= 23ohms.

Am I way off here?

Posted: Thu Feb 19, 2009 5:34 pm
by dave slagle
Welcome!

rather than reply to your actual numbers, I'll simply redo them.

Your 1:18 cinemags are called 150 ohm because the reflected load when terminated by the standard 47K will be 145 ohms. Add a little DCR and you are close enough to 150 for me. (47000/18^2)

So as it stands your cartridge sees 150 ohms and if you want to reduce that to 20 ohms by primary loading you will need to add 23 ohms across the primary.
(150*23)/(150+23)

I have not measured or seen measurements on the cinemags so i cannot really help you out with menno's method. Do you have measurement capabilities? You can do it by ear and that may be the "honest" way to do it so your ears will not be biased by the measurements.

If you want to try 20 ohms I would try the two extremes and see what you hear. The extremes are:

23 ohms primary 47K secondary
no primary and 6.5K secondary

both of these should present the same 20 ohms to the cartridge and if the transformer were ideal would sound the same. After listening to both for a while to get the feel, you can try the middle which would be something like a 50 ohm resistor on the primary and a 14K resistor on the secondary (leaving the 47K in place).

47K||14K= 10.8K (load resistor value)
10.8K/18^2= 33.3 (reflected load)
33.3||50=20 (load seen by cartridge)

dave

Posted: Fri Feb 20, 2009 4:31 pm
by Qman
Thanks Dave. Any advantage to loading the 1:36(40ohm) input instead. Also, do I need the R/C circuit added to the secondary for ringing?

Posted: Fri Feb 20, 2009 6:32 pm
by dave slagle
Again i am not really familiar with the trannies in question so the best I can say is try both ways and see. The math for the 1:36 is easy, just parallel the 40 ohm primary with a 40 ohm resistor to load the primary and simply add another 47K to se secondary to reflect back 20 ohms.

If you try the 1:36 you may want to be sure you are not over-driving your electronics.

as for the network on the secondary, you really need to look at each sitution on an individual basis. When you lower the cartridge impedance it tends to make the transformer ring but the only way to really know is to look at it on a scope. I am not a fan of using networks to fix a problem. If you know the source impedance and load conditions a transformer can be designed to behave "properly" and that is my approach.

dave

Posted: Fri Feb 20, 2009 8:27 pm
by Qman
So in both cases when loading the primary @ 40ohms or 150ohms I'm also going to load the secondary with 47K in parallel with my MM phonostage.

Posted: Fri Feb 20, 2009 9:30 pm
by dave slagle
no, your mm phono should already have 47K in place and you are sort of stuck with that load. if you add another 47K in parallel you then have a load of 23.5K which reflects back by the turns ratio squared

dave

Posted: Sun Dec 20, 2009 6:25 pm
by skudra
Hi Dave-I have a low output moving iron cartridge that likes to see a load of about 4-5k. Soundsmith Sussurro. I don't know it's internal impedance but I assume it's high.
If I understand it correctly, a SUT at 1:10 would show the cartridge 470 ohms which is too much loading. How then do we get the load to the desired value of ~4.5k?
And a 1:5 ratio would show the cart about 2k ohms, but the gain might not be enough?
Bill

Posted: Mon Dec 21, 2009 2:25 pm
by dave slagle
hey bill,

the sussurro has characteristics similar to the denon 103R. I'm not sure where you got the 4-5K number as maximum load since peter mentions it is around 1K. I know he was using a 1:10 which reflects back 470 ohms and didn't notice any HF loss in practice. I would think something like a 1:7 would be well suited and it will nicely present a 1K load and give you 17dB of gain.

dave

Posted: Mon Dec 21, 2009 3:02 pm
by skudra
Hey Dave,
I think I got one of the first ones Peter made, and he also made me his new firefly phono section that had a variable resistance control knob. He suggested 4.5 as being the right load at the time, but then I noticed recently that he modified it to 1k.
I think F.Schroeder likes even less loading-maybe even up to 47k.
We (A. Loesch) and some others had a cart shootout and ran the Sussurro wide open at 47k and it won the shootout hands down vs. a Zu Denon, Shelter 501, older Miyabi, Denon 301 and a few of Arthur's nos finds.
When I had the K&K phono stage, I felt that 1k was the cutoff, and anything more resistive than that lopped off the highs too much.

Posted: Mon Dec 21, 2009 10:10 pm
by dave slagle
hey,

I believe jeffrey has the boxed up SUT's that i did for that cartridge. He could get them up to you if you want to give them a try.

dave

Posted: Tue Dec 22, 2009 1:16 am
by skudra
Cool. I'll give him a call tomorrow.
Man, the SS Ultra High Comp Voice is mighty good. In most ways it betters the Sussurro and doesn't need a SUT. Amazing detail.

Posted: Tue Dec 22, 2009 3:07 pm
by jlsemrad
Dave, is there any advantage to loading the primary at the head shell?

John

Posted: Wed Dec 23, 2009 2:04 am
by dave slagle
Hey John,

To be honest I haven't thought of it. I can't think of a good reason one way would be any different than the other other than making the connections at the headshell being a PITA.

dave

103r

Posted: Sat Apr 07, 2012 2:01 pm
by deafbykhorns
It seems your familiar with the Denon 103r. What was the ideal turns ratio for this cartridge into a 47k mm stage?
I have an art audio phono stage with plenty of room inside that should allow mounting of an SUT
I think the Vinyl One used AN 1:10, secondary loading didn't seem to make any difference until it got to the point of reducing gain

Posted: Sun Apr 08, 2012 1:36 am
by dave slagle
First. Lets toss out that 47K number since it is meaningless for our purpose and it only limits our possibilities. For a MC or SUT input it simply shouldn't be there.

Without that resistor we are free to pick an appropriate turns ratio for overall system gain (remember to pay attention to headroom). Then the loading can be addressed via the primary and all is good.

dave

Posted: Sun Jul 29, 2012 2:34 am
by Maturin
Dave, let me see if I understand you aright or if I have it all bolluxed up: the best way to terminate a mc cartridge is at the primary of the SUT (or at the output of the cartridge), and the SUT should then terminate into a very high impedance input amplification stage?

I'm just trying to get the general principle correct, although I rather doubt I'll ever be able to actually afford a mc cartridge LOL.

VB, Maturin

Posted: Sat Mar 02, 2013 1:18 am
by dave slagle
Hey Maurtin,

You have it exactly right assuming the transformer doesn't go crazy into the high value load. Many transformers take advantage of secondary loading because it helps to quell ringing.

dave

Posted: Wed Aug 07, 2013 7:50 pm
by johs
Hello,
SUT loading to me really looks like a minefield. But it would be really foolish not to try primary loading when authorities do report improvment. To ensure a non authority like me;
Denon DL103 into Partridge 977 1:6 step-up ending in a Cursio Sarah with E88CC.
I have read in another forum that oscillaton may occour when the SUT is loading the input tube directly - are 88's in trouble here, would a 1Meg resistor help out if that's likely to happen? Any suggestion what value load resistor to use?
Regards, Johannes