I am lost here. Help! LOL.
I looked up RLC circuits and found how to calculate Xc reactive capacitance and XL reactive inductance, and impedance Z.
Xc= 1/(2*pi*F*C)
Xl = 2*pi*F*L
Z = R/sqrt(R^2 +X^2) where X is abs(Xc-XL)
Now I want to figure out how the signal is affected, going from the phono cartridge to the MC Step up, due to the cables R L C parameters across a wide frequency. How do I do this ? What is an appropriate model for this.
If I am using the Denon 103R, and a cinemag 3440 what values do I use and what equation do I put them into to find the voltage change at a given frequency.To my questions above I am guessing that: Once I know this I can change the frequencies for a given cable's C L R to find out the voltage change and convert it to decibles using -20Log (Vin /Vout) and it will become clear exactly what is happening when certain cables are used.
After I figure that out I would also be interested in calculating this wire affect for the connection out of the cinemag into a preamp's phono section. Then the connection from the pre to the power amp, and even amp to speakers would follow.
Perhaps the science will show that the selection of wire for those connections can make a difference within some parameters for R L C as they vary by cable makeup and length, and can help one in selecting an appropriate wire. for each part of the system.
Please help me out with modeling this as I am sadly a little lost despite my best efforts to try and figure this out. Thanks again.
|
I dont have this cartridge or step up, but I think it will be most useful to use it as an example since many people do.
Might as well post the specs.
Denon 103R Specs
Output 0.25 mV
Output impedance 14ohms
(no inductance spec, impedance is total I take it)
Cinemag 3440AH:
37.5,150,600 : 50K
(Into 47K rather than 50K this adjusts to 35 ohms, or 141 ohms depending on tap used)
Phono Section:
47K
Preamp:
Output Impedance 2K
(I figure 2k is good to use as most preamps will not be much higher than that value.)
|
Well, here is what I have done.
The computation does become complex, for spread sheeting etc..once you have to start looking into the differential equations. Do-able but much more practical to handle it using a brute force method, enter the circuit modelling software, which is what I did.
I used the circuit here used in this article: http://www.hagtech.com/loading.html
I adjusted the parameters for the 103R and a cinemag's reflected impedance. Here is the thing, for an MC setup, the capacitance and resistance parameters are of little consequence when the reflected impedance is used, contrary to the article.
Let me explain:
The article is looking at .47k 4.7 and 47k, but in reality the reflected impedance is .47K (470 Ohm) or less for just about any SUT. Now when examining the circuit at > 100 Ohms, minimum recommended loading for the cart we are talking about. It appears that the capacitance can create a huge roll off/tweak the subsonic spike as in the article. So this leads to looking at the assumed inductance. Most MCs have inductance between 5uH and 5mH ( 1mH =1000 uH). Now the results make sense, and rules of thumb/ common observations people have mentioned all come to light. The MC cartridge only becomes sensitive as the inductance moves higher.
For a cartdige with internal inductance 5-250uH, and a cable such as mogami 2549 which has 26pf/ft capacitance, 0.24uH/ft inductance and 0.022/ft Ohms resistance it becomes very difficult to change the subsonic spike by altering length and creating a roll off of even 0.1db at 20khz is just about out of question. The inductance is inconsequential all around, as for the capacitance well it depends.
So at the end of the article, they also explain this. The resonant qualities in a moving coil setup can essentially be reduced to its inductance relationship see . This I verified to be true when the carts output inductance is on the low end of the sliding scale. However as it approaches closer to 5mH this realtionship changes and it becomes very easy to create a less than optimal frequency response.
Final notes:
-Most available wires, used for phono interconnects and for tonearm wiring are suitable for the task and in MC setups do not create meaningful problems in response iff the MC cart's inernal inductance is low.
-As the MC cart's internal inductance approaches 5mh, all of these things become a problem, and it becomes easy to create a substantial roll off by altering interconnect length and by selecting different tone arm wiring. I.E/ The capacitance begins to matter MUCH more.
-Sadly the inductance spec is rarely given for any MC cartridge, so you are essentially in the dark before hand, but experimenting will reveal OBVIOUS results as the response curve can be very responsive when inductance is high, and will give almost no changes in the audible range when they are low.
-Low inductance MCs are clearly preferable and easier to manage within the system
-Loading for MMs, very different story, this can be a substantial problem. Capacitance does matter MUCH more.
-The circuit is simplified and does not give the entire story. The way that a phono stage responds to the ultrasonic information and the RLC can be complex, beyond the scope of a simple model and can be very specific depending on how your preamp operates. The pre/phono amp matters....DUH!.
|
Oh yeah, one more note on that article. It appears that they have grossly inflated the inductance figure on MCs, as I pointed out, for a typical MC cartridge. Some googling around can confirm that most are really a couple of uH to max a couple hundred uh MAX. However this was done to illustrate the point that they were trying to make, and their calculator is scaled back (but still inflated) to show a worse case scenario which is useful to just about anyone out there using the calculator with modern MC.
|
I can say I only partly understand, I will have to learn a little more to fully wrap my head around how it works. However, for now:
I tried modeling the transformer, not sure why I am running into problems here. I will try to post the image later maybe you can offer some guidance on the schematic/model for simulation. again ,much appreciated thanks guys! |
AL, thanks for pointing that out!!!!
You are right I must have not typed in uh or mh and it defaulted to full henries, and I completely missed this. I will have to look at this later when I get a chance.
But now that, that is a thing I might as well figure out what the inductance is in acutal step up transformers. I emailed cinemag hopefully they respond. |
Jcarr, thank you very much for posting I appreciate it. I will read through the forum, gather my thoughts, and post in this thread again. I am also wondering, what do you find is the typical inductance of a moving coil cartridge (low output and also high output) ? I get that several generators can be used but I am just curious what your take is on this. Are we in the realm of 5-500uH or are we more in the area of 5-500mH?
I can see what you are saying. I will put it into practical terms for anyone else who is reading so they can follow as well ( all 2 of them). Suppose that you have 100pf cable from the cart -> Sut and another identical one from the SUT->Preamp. and you are using a 1:10 SUT. Then...
The cart sees:
-100Pf + 100pf *sqrt(10)= 416pf
-The 47K load gets reflected back at 470 Ohms
-and the inductance of the wire is meaningless even after -being reflected with a Sqrt(10) multiple tied to it. The -cartdrige inductance is x * sqrt(10) and can be substantial depending.
The SUT sees ( bad phrasing, Through the SUT...):
-100pf *10 +100 pf = 1,100pf
-inductance from the cart x*10 = 10X
-and 40 Ohms from the cart *10 + 47K = 47400, or just 47K |
Al, thanks for pointing out those details. You are right. I had not looked at the too little side. On the correlation I would hope so, it would make things easier but there in solid convention for quotation that would make it easy to figure out, but I will have to think about this. I do see what you mean, X MV output and Y resistance into an open circuit figure out ballpark for inductance etc.
As for the Ultrasonic Subsonic...tomato tom-ah-to LOL, you are correct I goofed on that. |
Jcarr, Please email the white paper, you reference in the thread I am exceedingly interested in reading it!!!! Thanks. I will PM you. |
Just realized your user is set to, Private and I cannot PM you. |
Wait a Minute I had a dull moment let us redo this:
Too bad there are no edits, now the mistakes look silly forever.
Assumptions/Signal Chain
Cart[x uh,30Ohms , wire(100pf,2uh,1 Ohm)->SUT->(Same wire)->Preamp(47K)
Cart sees everything on its side and the other side of the SUT DIVIDED by 10^2 or 100:
-100Pf (from the Cart->SUT connection) + 100pr/(10^2) (From the SUT->Preamp) back through the SUT = 101pf
-The 47K load gets reflected back at 47K/(10^2) = 470 Ohms
-and the inductance of the wire is meaningless and even less so after it is divided by 10^2.
-The cartdrige inductance is x and can be substantial depending as explained earlier
Other side ofSUT sees everything stepped up by 10^2:
-100pf *10^2 +100 pf = 10,100pf (WOWZER)
-inductance from the cart x uh*10^2 = 100X uh
-and 40 Ohms from the cart *10^2 = 4K
driven into a pre with a input imp of 47K.
This should be correct, sorry for the mixup, and please let me know if I made a mistake here (again).
Note: I have ignored reactance (Capacitive and Inductive) though I do not think it is substantial and I am also not certain how that follows through the ends of a SUT, If I were to guess Z is altered as sqrt((Xc/10^2)^2 + (XL/10^2)^2) using a linear transform of variable...ditto for the other direction. |
Jcarr, Maybe I am off here but on the primary side/ what the cartridge sees: Do you not divide by turns ratio square i.e/ 10^2 or 100 in my example.
So:
47K becomes 470 Ohms
100 pf SUT to Preamp becomes 1 Pf
??
I would have thought that this implies that the connection from the cartridge to the SUT may be more capacitance critical than the connection from the SUT to the pre.
going to the secondary side however requires multiplication by turns ^2 or by 100.
Please let me know what I am missing here. |
Hi guys, I have been having a lot of trouble modeling this correctly with the SUT in the mix. I am sure I am making some silly mistakes but I can't find them. Here is a link to the image of the simulation + circuit model. I tried to follow the model from the snippets I could find online to the paper by J carr, I would love to read the whole paper one day..please PM me since I cant PM you Jcarr.
For the earlier simulations it was easy to lean on the Hagerman article and simply copy the circuit and play with the parameters. But, in this case it is not so simple, and I would appreciate some help. Thanks.
http://s27.postimg.org/aage8hg2/Attempt_at_Circuit_Model.png
I am assuming the same cart specs from the paper, and using the specs for the mogami 2549 wire, 0.022 Ohms/foot, 26pf/foot, and 0.72uH. I used a 1:10 Step up ratio for the SUT.
Please tell me how if I have made a mistake here ( I am sure I have), and how to correct it as I am interested in seeing this display something meaningful. Please be gentle with the commentary as I am a rookie and know nothing about Electrical Engineering. Thanks again. |
Link was not working lets try again:
http://s27.postimg.org/aage8hg2r/Attempt_at_Circuit_Model.png
http://postimg.org/image/6e32chv33
http://postimg.org/image/6e32chv33/full
Hopefully one of those links works. |
agreed John. Once the model is set. However once it is set is easy to simulate with software which is a more flexible and visual way of observing the data.
The thing is, the basic model is not complete. Have you looked at the links above to see how it was modeled? Does this make sense ? |
Thanks Almarg, for you insight.
as to the items:
1) In the K statement I have modeled it as a perfect SUT, this can be seen the by the 1 in the " K L2 L4 1". I can use numbers of less than one to make up for losses due to the SUT, like .6 .7 .9 etc..I figured lets see what happens with a perfect transformer first to isolate the cable effects.
2) Good point, I will sort that once I have the rest of the model making sense. I am concerned that the whole network of connections is completely modeled incorrectly. Can you please give some input on the basic structure, thanks.
3)Derived to have a 1:10 ratio, no other "genius" there (Strong emphasis on the quotation marks). The values I chose were arbitrary, so long as they had at square of quotient to give 10. Maybe that is incorrect.
4) I hear you, but I want to make sure that model itself I.E/ series/parallel resistors ground inductors and everything else are placed correctly in the circuit to model that part of a stereo system. I tried to follow Jcarr's model, but in the end I am still not sure why when modeling a wire connecting devices you would model the RCL in series or in parallel, and in what order you place the RCL for the wire connectors. I am a little lost on it and hoping someone can look at the images of the model circuit I have done and help me tweak it to the one that is correct.
|
Typo: Under item "4) ...place the RCL for the wire connectors.."
I do not mean wire connectors, I mean wires or interconnects. |
Hi,
According to this link the:
"...Remember,
the inductance is proportional to the
square of the turns ratio. In the ex ample above, a turns ratio of 1:3 gives a 1:9 inductance ratio..."
they they proceed to use 100uH : 900 uH.
I have not even got the slightest clue where to get the actual figures for the SUTs, as none of them publish this information. I suppose I can try to email cinemag, Jensen etc... to see if I can get it.
http://cds.linear.com/docs/en/lt-journal/LTMag-V16N3-23-LTspice_Transformers-MikeEngelhardt.pdf
Now, I followed this guide to implement the SUT. Worth repeating: I know very VERY little about electrical engineering. However, in the current model as I raise Capacitance massively on either side of the SUT is makes NO difference to the frequency plot. Also the Freq plot is offside by over -50db...as it should be reading +20 after going through the SUT. Hence Why I think there may still be something off. I dont know enough EE to figure out what the "something" is. Please experts, lend a helping hand, CALLING ALL EXPERTS PLEASE!
It would be great if I could attach the model so you can open it in your own software without re-sketching it. If you know of a way please tell me, I cant seem to attach files on audiogon. |
I tweaked the model, after reading another paper on tranmission lines etc.. I found a model that made sense and demonstrated the correct db gain and presented a familiar looking bode plot. I think this lumped model may be correct, or at least "better".
See the link
http://postimg.org/image/dgyazcfhj/
I tested this model with some extreme values, and compared to the baseline case outlined in Jcars white paper. I used the specs from the Mogami 2549 out to as much as 20 feet worth of RCL, tried changing the cartrdige parameters for impedance up to 50 Ohms and 500 Uh etc...
The only changes I have noticed in the response curve are in the ultrasonic peak shape and neighborhood, though this activity occurred far away from the audible band (100K or more). The Subsonic roll off was very slightly affected and was still very low (<15Hz) across the values used. Of course this is under the "Perfect" transformer assumption. So this model is showing that a cable with the characteristics of the 2549 is capable of reproducing the signal without an drastic changes to the frequency response in an MC Phono rig, provided that the parameters of the cable are "reasonable" and inline with what most people would use anyways. |
Lewm, can you please give a link. I have looked on Jensen and not seen these specs. Thanks. |
Looks like so long as L4 is at least 100mh things look normal. The pairwise inductance have been chosen to give a 1:10 ratio i.e/ as per the source it is the sqrt of the inductance ratio. (L4,L5)->(1uH, 100uH) ; (10uH, 1mH) ; (100uh, 10mh) ; ( 1mH, 100 mH) ; (10mH, 1000mH) were all giving weird results as in the first circuit and the gain was below the approximate 20 Db I was expecting as per the normal calculation given the turns ratio.
Once I used values in the form of ( X, 100X) where X was greater than 100mh things normalized and stayed normal. I tried (100mH, 10000mH) all the way to (10H,1000H) the gain stayed constant, the low freq roll-off disappeared as the values got pumped up, though it makes little difference as much of this activity was below 20 HZ. The ultrasonic peak remained above 100K and well outside of the audible range, its shape was slightly altered as the values went higher.
So as it sits now. I am very curious know exactly what an actual spec for the prm/sec inductance values are in transformers that we use. NO manufacturer has this spec on their website that I have been able to locate, so hopefully cinemag emails me back. I will report back once I know OR if anyone else has a reliable source for these specs please post. |
Emailed both Jensen and Cinemag, no word yet. Thanks for the info Lewm ! |
