Capacitance / resistor loading question

OK, I've tried using online parallel resistor calculators and I can't make heads or tails of them. I've never been a numbers guy. Can anyone help me out here?

I have an early '70s vintage receiver with 47 kOhm input resistance, so exactly what resistor value do I need to get a 68 kOhm load? (This is so I can get better treble response from a Shure M97xE.)

Thanks to anyone who can answer this!

 

Unfortunately you can'r use parallel plugs to increase resistance. Resistance in parallel is always lower than the lowest value, so no matter how high the resistance you add, your receiver's 47kohm will cause the total resistance load to be less than 47kohm.

 

You might check over at vinylengine.com or at diyaudio.com or audiokarma.org forums for advice from folks who've had success with the M97xE (I recall there's an Australian poster at one of those forums who's a big fan of this cartridge after modding it). Even with tweaking my understand is that the M97xE is a warm sounding cartridge. I have seen some folks online swear by this cart combined a JICO stylus.

I just did a quick check of its specs with the Cartridge/Amp Loading/Resonance calculator at http://www.vinylengine.com/cartridge_database_tools.php. With its 650mh inductance, you need about 290-295pf of capacitance to get the optimal loading value to 47kohms. Capacitance is the opposite of resistance in that combining capacitance in parallel increase the total capacitance. Of course you need to know your receiver's capacitance rating, which may or may not be available. Plus you need to consider the wiring in your turntable an arm plus your interconnects. That can run about 100-150pf (or more if the interconnect is cheap).

 

Hmm, that may explain my difficulty. I was going off this advice from Antares in the needledrop thread:

The thing is, I've looked elsewhere online and I've seen other people also say that there's a way to use this method to boost the M97xE's highs. The Shure recommendation for this cart is 250pF.

 

We crossposted! Thank you, Stefan.

 

As I understand it with cartridge loading what you're really doing with the capacitance and resistance loading is tuning the resonant frequency of the harmonic oscillator formed by the inductance of the cartridge, the capacitance of the cable and the phono input, and the impedance of the phono input.

You can't change the inductance of the cartridge, so you're moving around the resistance and capacitance to get the frequency response you want without undue ringing.

The capacitance loading is going to tune the frequency of the resonance -- more capacitance will move the frequency lower, less capacitance will move the frequency higher.

The resistance loading will damp the amplitude of the resonance, so higher impedance values will knock down the height of the resonant spike, lower values will allow the peak to be higher.

With the cart's 650 mH inductance, at the manufacture's recommened loading of 200 pF and 47K ohms according to this calculator (http://www.calctool.org/CALC/eng/electronics/RLC_circuit) you'd have a peak at about 14 kHz, which means the high frequencies would start rising somewhere around an octave below the peak, say around 7 khz, rise steeply to 14 khz, and fall steeply after that.

That's far from flat obviously through the entire "presence" band of upper treble (with a pretty steep roll off of the very top end, although I for one, at 48 years of age, don't have any hearing to speak of above 15khz anyway). But usually a cartridge manufacturer will design the cart to exploit the mechanical tip resonance such that at the recommended electrical loading the actual frequency response is flatter and more extended at the top end than what the electrical model of the circuit would suggest would be the case at the recommended loading.

So theoretically you might want a lower capacitance load to push the resonant frequency above 20khz, and then the requisite impedance load to damp the spike so that frequency extension is flat out through the range of human hearing, but if the manufacturer has designed the cartridge around a mechancial resonance, that kind of loading might not be ideal. My guess is that the loading Shure specifies gives the flattest response.

 

Yeah. It's not bad as it stands, just wanted to see what a tweak might do.

Unfortunately, nowhere on the web can I find the capacitance for my Sylvania RQ3748 anyway, so it's all kind of moot.

 

Actually in my experience and according to info at Hagerman Technology LLC: Cartridge Loading, the opposite is true.

Lowering the resistance dampens the resonance spike more while higher values dampen it less. When I first started experimenting with resistance loading, I built a little box with a stereo potentiometer connected to two cables with RCA plugs, which I plugged into my Y-adapters. While playing white or pink noise from a test record and monitoring the frequency curve using Adobe Audition in real time, it was easy to see that lowering the resistance tamed the resonance peak. The only problem was that it never completely flattened the peak before cutting into the flat surrounding frequencies and dipping them. It was sort of like running a parametric EQ with a boost at a high Q value then adding a cut at the same frequency but with a lower Q value.

There's also a physical resonance related to the tip mass. I see discussions of that quite a bit on the other forums I mentioned. I also saw a big thread last year on vinylengine where someone tried achieving flat frequency response by adding inductance in series (IIRC).

 

Sorry Ben, I steered you in the wrong direction. I've used this method to reduce brightness with an AT cartridge, but not the other way around and didn't think far enough ahead. Just like you can't reduce the capacitance of a phono stage externally with loading plugs, you can't increase the impedance without internal surgery. So short of opening up the receiver and looking for the phono input's resistor, there's no way to do this I'm afraid. If there's a 47k resistor in there, it would have to be replaced with a 68k version or put in series with 21k to obtain 68k.

 

Yes you are correct, a lower value of resistor say 1000 ohms vs 47000 ohms would damp the amplitude of the peak more, if I wrote the opposite above it was a brain cramp.

 

What I don't understand is, if there were a way to fine tune any given cartridge by adjusting capacitance and/or resistance in order to achieve a better frequency response, why wouldn't the manufacturer incorporate it into the design? (Or, of course, the recommended loading.) That makes me wonder if all attempts to do so are self-defeating.

However, fine tuning to suit personal preference is another matter entirely.

 

Well, I suspect most manufacturers do. When Shure recommends for the OP's cart, 200-300 pF capacitance and 47k ohm impedance, my guess is that those recommendations reflect the loading that they've designed around for the flattest frequency response for that cartridge using typical real world phono preamp circuits.

 

There's an interesting review of the M97xe at http://www.tnt-audio.com/sorgenti/shure_m97xe_e.html

The author's discussion of recommended capacitance and his frequency graphs are notable.

Here's his graph for 250pf loading:

Notice the dip around 5-7kHz. This is an area that really affects the perceived brightness of a cartridge. In fact, it reminds me what I wrote earlier today about how lowering the loading resistance too much tended to cause a dip in the areas surrounding the resonant peak.

If you go to the vinylengine.com loading calculator and enter 47kΩ, 650mH (the M97xe's inductance) and 250pF then click the Calculate button, the resulting Resonance peak comes out at 12.5kHz, which looks about right if you discount the little bumps and dips in the line. The optimal loading is 51kΩ, thus a standard 47kΩ is too low.

Now look at his test graphs for various capacitance values:

Match these with the vinylengine.com calculator and you get the following results for Resonance peak and Optimal load:

150pF: 16.1, 65.8kΩ
250pF: 12.5, 51kΩ
370pF: 10.3, 41.9kΩ
620pF: 7.9, 32.4kΩ

The capacitance I mentioned earlier today of 290-295pF would result in a peak at 11.5 that would be best damped with an optimal loading value of exactly 47kΩ! Not that this would produce a spectrum somewhere between the red and blue graph lines, which would result in a slightly brighter sound in the upper midrange/lower treble. It'd roll off the upper highs, but that wouldn't affect the perception of "liveliness" as much as the slight boost around 5-7kHz.

 

It is well documented that the Shure M97xE has a sloping treble response with today's typical loading settings, so if that were Shure's intention, it can not have been to obtain neutral playback. It is also my understanding that 47 kOhm is more an historically evolved/accepted value than an actual standard. With the Shure's basic design going back to a time when 47k was even less "standardized" (many amps of those days had 60-75 or even 100k settings), I do not believe that the current frequency response at 47k was in Shure's original design premise. You can always explain it now by saying that it would appeal to people who prefer a "softer" presentation or that it reduces playback noise/clicks better, but that's just explanations after the fact imo.

So, with the luxury of a push-button adjustment, is your Shure V15/SAS too bright at 68k?

 

Just looking at those frequency response plots, which show a steady 3-4 dB rise from around 10khz to around 200 hz regardless of the loading, it seems to me that, yeah, you can play around with the top octave response via loading changes but the cartridge is probably always going to have a recessed sounding high end and a prominent lower midrange.

 

I've found the most reliable way to determine cartridge loading is NOT using calculations. I use The Ultimate Analogue LP and run the 3 high frequency tests (1kHz and 10kHz reference tones and a 1kHz-20kHz sweep) and record it into my computer. I look at the waveform and the dB meter and make electrical adjustments to get the flattest response possible. It'll never be totally flat, but I can usually get carts within ±2dB.
For example, a Shure V15VxMR works best at the recommended 200pF/47k, but with the Jico SAS stylus it needs to be loaded at 570pF/22k to get a similar response. At these settings, the MR stylus has up to a maximum -2dB sag at about 8,000Hz, but is flat from about 11,000 to 20,000Hz. The Jico SAS has a maximum -1.5dB sag at about 8,000 Hz, is flat at 10,000 Hz, is at +1.5dB peak at about 15,000Hz and slopes down to flat at 20,000Hz.
An Audio Technica AT150MLX says to load it at 100-200pF/47k. Its measures terribly and sounds horrible at this setting. If loaded at 570pF/15k it sounds wonderful. With that setting, It has a maximum -1.5dB sag at about 8,000Hz; +1dB at 15,000Hz; and -1dB at 20,000Hz.
I did a A/B CD/Vinyl test with the DCC version of Van Halen S/T as well as the MoFi version of Beck's Sea Change. At the above settings, the LP's sounded nearly identical to the CDs. Any other setting on my pre amp (EAR 324) didn't work as well as the above settings I got from using the Test LP.
I used to think that I was hearing a lot of extra resolution with vinyl. It turns out that the recommended cartridge loading sometimes would produce a response of +7dB at 15,000Hz! All that "air" was really just frying my cilia, fatiguing my ears, and highlighting surface noise. Properly loaded, now I can hear what's really on the vinyl, vocals sound more natural, less surface noise, and a more balanced sound. I've found that even if the tonality closely matches the same CD, an analogue pressed record can still have more front to back depth of sound.
As aside note, the total capacitance ratings listed above is the setting on the pre-amp plus the 100pF of my interconnect.

 

An exceptional post. It seems to combine the best of the empirical approach with actual listening experience.

It would be interesting if someone could replicate your results, to confirm them. I'd almost volunteer myself out of curiosity except I've no idea how to do it and would worry about the time it would take.

One thing that strikes me is that you have to make quite drastic alterations in loading to achieve noticeable changes in outcomes. To answer Antares's question, I couldn't detect a difference when switching between 47k and 68k on my Shure V15VxMR + Jico SAS (capacitance unchanged) and now I know why - from what I've learned here it would probably be inaudible during casual listening, even if just measurable using electronic equipment.

Another thing I suppose is that another person's cartridge/stylus assembly might yield different results, and the arm might make quite a difference too, which is why I think one would need more than one set of results before leaping to any conclusions.

However I think those results are very interesting indeed and I'm very tempted to try out your AT150MLX settings on my old AT440MLa. Oh yes, and I meant to ask, how far off from flat was the Shure V15VxMR + Jico SAS WITHOUT the adjustments to loading?

 

I just got my first ever phono pre-amp, having always had an ancient amplifier, and it has a capacitance selector (100 or 200pf) and gain control. Two capacitance choices is a bonus, but until this thread, I had no idea what the hell it did. My cart is an AT311EP: Frequency response: 15 to 27,000Hz, Vertical tracking force: 1.0 to 1.5g, Channel separation: 29dB, Output: 5mV, Vertical tracking angle: 20 degrees, Stylus: 0.3 x 0.7 mil BiRadial, Cantilever: thin wall, Load impedance: 47K ohms.

Assuming all AT's are bright, am I correct in assuming I should be going with the 100pf setting?

One review of the pre said the gain control can be a little "tweaky". How does one set it for the most balanced frequency responce, rather than just how loud you want it?

What other info/variables/numbers should I be taking into account? Never thought I would become an amateur electrician when I plugged my TT back in. If I am hijacking this thread, I apologize and can start a new one.

 

kinkling, welcome to the discussion. No need to start another thread.

The inductance for the AT311EP isn't readily available through a Google but I found an indirect reference that seems to indicate it's from a family of AT cartridges that are all around 420mH. This means that, according to the loading calculators at vinylengine and hagerman, a total capacitance of 190pf would result in a resonance at 17.8kHz with optimal loading for damping that resonance at 47k. Given that your turntable and cable to to the phono pre-map (interconnect) will run you at least 75-150pf, you'll probably get the best sound with the 100pf setting. Raising the capacitance means the resonant frequency comes down into the more audible ranges and consequently to tame the resonance, you have to lower the loading. For example, if youre wiring/interconnect gives you 150pf and you use the 200pf setting on the pre-amp, you'll end up with the resonance at 13kHz and required loading at 34.6kohms.

The problem I've noticed with my experiments over the years with loading situations like this is that the whole damping approach may tame the resonance but it also rolls off the highs above the resonance. For example, if you look at the graph for which I posted a link to earlier in the thread:


Notice the yellow line for the 620pF loading. This was for a Shure M97xe, which with its 650mH inductance, and a 620pF capacitance results in a resonance around 7.9kHz. That was from the vinylengine calculator and the graph's yellow line indeed peaks around 7.5-7.9khz (remember these graphs were done with pink noise so they are always slanted downwards towards the right). Note how steeply the high frequencies drop off above the resonance. This is what happens with a high capacitance. It moves the resonant frequency downwards. As the Hagerman loading site points out, yes, you can damp this with loading but that won't prevent rolling off the frequencies above the resonance. Some folks mistake this rolling off for being "warm" but other just find it dull.

So the best strategy for a moving magnet is to keep the resonant frequency as high as possible and keep the loading as close to the optimal as possible. Too high and you'll get a dip in the upper midrange; too low and you'll roll off the highs.

This is the theory, but through my last four cartridges, I've found the calculations and adjustments to be spot on. When my capacitance was too high, the upper midrange/lower treble was bright and brassy but there were no extended highs. Once I got the capacitance down to where the resonant frequency was higher, things balanced out.

In the end, you have to come up with a sound you enjoy and that's entirely subjective. So experiment with both settings on the pre-amp and see which gives you the sound you prefer.

 

Thanks, Stefan--as far as what I preferred, of course I could not make out any difference whatsoever (too many Motorhead shows, perhaps?), but I hadn't brought any files into the computer to scope them out and see what was going on, either. How would the pre-amp's gain control affect frequency response?

 

Gain will (ideally) not affect frequency response, kinkling. It can affect perceived dynamics of the music though. Too little gain can make it sound flat and lifeless, too much and things can get pushy and compressed sounding.

Gain is not directly related to the electrical resonance being discussed here. Moving Magnet cartridges like your Audio Technica rely on an electrical resonance for their treble delivery, which is determined by the resistance and capacitance values in the circuit formed by the cartridge and phono-pre input stage. Resistance determines the amplitude of the resonance peak, capacitance defines the frequency at which the peak occurs. That's the (electrical) theory. Cartridges being electro-mechanical devices, there are other considerations to take into account, but this electronic resonance will play a major role.

 

THAT was precisely the concise explanation I have been trying to find for weeks...thanks very much!

 

I just made some loading plugs this morning, and am listening at the moment. I don’t have the best gear for recording results, but I'll try to get a pink noise graph up soon.

 

AT150 MLX @ 21K/526pF, Pink Noise

 

View attachment 240637 I haven't read all of the posts in this thread, but I just did the Ultimate Analogue Test LP high frequency test with a brand new M97xE through an EAR 324 phono stage and found that it does not have a roll off in the treble at all. It doesn't sound that way either. My tonearm wire adds 100pF capacitance. The picture I'm posting shows the left sine wave as a 1kHz test tone, the middle sine wave is a 10kHz tone that should aim to be the same amplitude as the 1kHz, and the sine wave on the right is a 1kHz-20kHz frequency sweep. To get a sense of scale, the difference between the 1kHz tone on the left and the 10kHz tone in the middle shown on the test on the top of the picture is approximately +0.5dB to +1dB. The test results at the top is loaded at 33kΩ impedance/570pF (total capacitance) and is the flattest response I could achieve with nontraditional loading . It sounds dead neutral when comparing CDs and vinyl with the same mastering (DCC Van Halen, etc). The bottom test is a more traditional loading at 47kΩ impedance/200pF (total capacitance). It still sounds really, really close.
Hopefully, this will prove that the M97xE is NOT dull/rolled off/etc. I have found that most other cartridges that test bright sound bright. The only other cartridge that tested this flat was a Dynavector 17D3. This Shure cartridge is simply amazing. It was aligned using the Stevenson geometry on a Mint LP Best Tractor. I highly recommend buying the Ultimate Analogue Test LP, needle dropping the tests, looking at the waveforms' amplitude to determine the proper loading for your particular system to achieve the best results. It's worked every time for me. Hope this helps!