Owner's Hot-Mods - Variable phono input loading

Print Page | Close Window

Variable phono input loading

Printed From: Graham Slee Hifi System Components
Category: DIY AUDIO
Forum Name: Owner's Hot-Mods
Forum Description: Tell us how and why you hot-modded your audio gear
URL: https://www.hifisystemcomponents.com/forum/forum_posts.asp?TID=964
Printed Date: 27 Mar 2026 at 2:12am
Software Version: Web Wiz Forums 12.01 - http://www.webwizforums.com

Topic: Variable phono input loading

Posted By: Fatmangolf
Subject: Variable phono input loading
Date Posted: 19 Aug 2010 at 9:58am

In the interest of experimentation I am going to develop a way to safely change the input capacitance of my Genera without switching it off. I have written elsewhere about modelling the electrical circuit, this thread is more practical.

I appreciate the following:

The sound will change over time after a "cold" capacitor is added
A cheap and nasty capacitor will sound bad
The impact of changing the LC tank and the damping resistor is only part of the HF response of the cartridge
The input capacitor also helps to stabilise the preamp (please see Graham's work on the Genera)
The Spice model suggests the resonant peak and roll-off changes slightly
And yes this is like nudging the treble control

But I am still curious. My thinking is to use snipped pieces of component lead or tinned wire through the board, solder across a smaller capacitor i.e. 50pF and 2 old cartridge connectors or sockets from a D-SUB connector. The extra capacitor would plug into this.

Before I get the tools out has anyone any practical tips from trying this (or similar mods) themselves?

Jon

-------------
Jon

Open mind and ears whilst owning GSP Genera, Accession M, Accession MC, Elevator EXP, Solo ULDE, Proprius amps, Cusat50 cables, Lautus digital cable, Spatia cables and links, and a Majestic DAC.

Replies:

Posted By: Graham Slee
Date Posted: 20 Aug 2010 at 2:32am

Suggestion...

Some stripboard (veroboard), a DIL switch and the selection of caps you wish to try - one end to DIL switch the other to ground - the DIL switch commons to the input.

Correction...

The input cap does not help stabilise the amp - it is stable without. It prevents the input slew rate being faster than the output slew rate to combat slew rate distortion.

-------------
That none should be able to park up and enjoy the view without a smartphone and the knowledge in how to use apps

Posted By: Fatmangolf
Date Posted: 20 Aug 2010 at 7:58am

Great, that would be a very neat and reversible approach. Correction noted. Thank you.

Posted By: Fatmangolf
Date Posted: 21 Aug 2010 at 12:07pm

The%20input%20cap%20switching%20from%20above

With an early start the Genera was back on for a quick test by 9:30am. All seems to be fine but I want it to have some running time after being switched off so I have left all 4 capacitors switched on in parallel.

Even with Graham's helpful suggestion you can see the extension boards aren't pretty. It is more robust than the approach I had considered where the leads would have probably wobbled out! Anyway these extensions will allow me to try different capacitor loading when the preamp has warmed up. Even now the differences can be heard on percussion sounds.

(I deliberately did not re-use the original 220pF capacitors on these extension boards so I can easily reinstate the original PCB/circuit)

Posted By: Fatmangolf
Date Posted: 21 Aug 2010 at 10:43pm

This is why I wanted to experiment with the actual sound of my Goldring 2500 and the Genera. After some issues with my Rega Exact (low self-inductance for MM), I tried the Golrding 1012 (550mH) and was tempted to buy a 2500 (720mH), which I would recommend trying with a Rega arm.

Electrical%20response%20for%20different%20input%20capacitors

The curve shows a Spice model of the circuit described elsewhere in the Forum, specifically the electrical parameters of the cartridge (dc resistance and self inductance) and their interaction with the input capacitance and resistance of the preamp.

You can see I have kept the input resistance at 47K and modelled the effect of different preamp capacitors. The cable capacitance is about 100pF, so 100pF above gives a combined load of 200pF. To my inexperienced eye the phase and other curves seemed quite similar so I have just used the response curves above.

You can see the more extreme effect of having all the DIL switches on with the 570pF curve's peak, that is just as dramatic in real life as in Spice! I made the modification to hear the quite subtle difference between the "stock" 220pF and the 100pF. I prefer 100pF with the Goldring 2500 but probably the other way round with the 1012, although I did add a damping resistor to tame its treble.

After all this I think the 50pF and 100pF sound best! Hopefully someone else at some time will find this helpful.

Jon

Posted By: Graham Slee
Date Posted: 22 Aug 2010 at 12:42am

A comment: I find the Evox PFR polypropylene we use for the 220p input cap (as well as in the EQ) differ considerably in sound to polystyrene. If you were to solder a couple of bits of wire to the original 220p cap and compare it against the 220p polystyrene and let us know your findings?

-------------
That none should be able to park up and enjoy the view without a smartphone and the knowledge in how to use apps

Posted By: Fatmangolf
Date Posted: 22 Aug 2010 at 9:43pm

To be fair the batch of polystyrenes were used because 1) I could get them easily and 2) they would all be the same age and 3) the comparison would be between the pF values, not the compositions. I would definitely have used a 50pF Evox PFR instead if I had one!

I will do the 220pF listening comparison you suggested and send you a note. I will try a blind test as well, don't expect a surprise win for the p/s if the first impression was anything to go by! The new P/S 220pF did not compare well with the 2 month old Evox 220pF, I could hear why you chose the latter component.

Graham, where can I order Evox PFR 50pF or 100pF from please? Selecting between one of those and the 220pF that came with the kit would have been my ideal test of the LC tank for my cartridge.

Posted By: Graham Slee
Date Posted: 23 Aug 2010 at 10:06pm

Check this link: http://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=2404707

You'd have to do a bit of cheating to get through the registration process (trade only... Wink

) and then you can order online (like most do... Wink

), or if you know somebody...

By the way, 100pf is the smallest value in the range.

-------------
That none should be able to park up and enjoy the view without a smartphone and the knowledge in how to use apps

Posted By: Fatmangolf
Date Posted: 23 Aug 2010 at 10:12pm

As agreed I did a quick snip and solder to put the polystyrene 220pF and Evox 220pF next to each other, then left both connected to op amp input and +9V potential for a few hours. This is an early test.

The Evox is better, it is not about the sound being brighter or slightly harsh on the polystyrene. I think the Evox is easier on the ear. I would normally describe the difference as detail or resolution but it is probably speed of response to transients. The mix is clearer with instruments more distinct and reverberation/delays being audible.

The best example I can give is a hi-hat. With the Evox, the stick hits and foot work can be heard in the mix. The polystyrene rendering makes it more like a ride cymbal, splashier for want of a better word. I am surprised how big a difference this makes.

Both 220pF's are a little trebly with my Goldring cartridge. But choosing between them is no contest! How many years ago did you discover these little gems?

Posted By: Fatmangolf
Date Posted: 23 Aug 2010 at 10:13pm

Graham Slee wrote:

"Check this link: http://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=2404707

You'd have to do a bit of cheating to get through the registration process (trade only... Wink

) and then you can order online (like most do... Wink

), or if you know somebody...

By the way, 100pf is the smallest value in the range."

Thank you Graham.

Posted By: Graham Slee
Date Posted: 23 Aug 2010 at 11:44pm

"How many years ago did you discover these little gems?"

10!

-------------
That none should be able to park up and enjoy the view without a smartphone and the knowledge in how to use apps

Posted By: Graham Slee
Date Posted: 24 Aug 2010 at 12:22am

Just to clarify because I did get a bit lost for a moment, the curves are thus...

(looking at the right hand end of the chart)

The biggest peak +3.5dB is the pink line into 570pf load (is this the total load, i.e. 100pf wiring plus 470pf cap?)

The next biggest peak +1dB is with the standard 220pf load cap (in parallel with the arm wiring - total 320pf)

The flattest almost as extended response is the 100pf load cap (plus the 100pf arm wiring making 200pf)

The early roll-off - the lowest curve - is with 0pf (plus arm wiring equals 100pf)

Generally, the correct total loading is 200pf for a typical 500mH magnetic cartridge. However, that can cause (to use an old carburettor tuning term) "hunting" where the input slew rate can, in some circumstances, lead or match the phono stage output slew rate.

The 220pf load cap plus 100pf arm wiring is a case of finding some common ground - limiting input slew rate (only slightly though), but with a +1dB peaking around 16kHz (if memory serves me...)

My old favourite was always 100pf (plus 100pf wiring) but over the past year I've felt the 220pf (plus 100pf wiring) better for my tastes. It should also be noted that arm wiring capacitance can vary +50%/- 30% depending on manufacturer (in my experience), and on the Hadcock Integra tone arm... well, I shall say nothing much apart from the peak will go off the Y-axis - but advise that the standard model (with the pig-tails) is better and the user to get a 1.5 metre Cusat50 (sorry about the "plug").

In passing I shall mention the popularity of the Ortofon 220pf adapter plug back in the late 70s. Tone controls had been "abolished" by then... Cool

(with the 220pf adapter plug, a DIN standard onboard phono stage with 220pf loading cap and 100pf arm wiring equaling 540pf there must have been some dull sounding phono inputs back then... Wink

)

-------------
That none should be able to park up and enjoy the view without a smartphone and the knowledge in how to use apps

Posted By: Fatmangolf
Date Posted: 24 Aug 2010 at 8:14am

"The biggest peak +3.5dB is the pink line into 570pf load (is this the total load, i.e. 100pf wiring plus 470pf cap?)" Yes it is, and your reading of the other curves is spot on.

The Goldring 2500 I am now using has a higher inductance of 720mH so you will already realise the LC curves are different from the "standard" 500mH. It peaks at 10.5KHz, lifting 5 - 10kHz like a treble control, and rolls off lower that a 500mH would. For this I am using the formula, Fpeak = 1 / (2 x Pi x square root (L x C))

Goldring recommend 100-200pF total loading, including the cable. Anyway I was curious enough to experiment as described above.

My cartridge and turntable are not exotic or high end, but I know what you mean about the 220pF plus cable loading. That gave a nice emphasis and definition with the Rega Exact. I have experimented here to fine tune it for the Goldring 2500 but it definitely needs the right parts! You were right, of course, about my comparing the 220pF PS with the Evox PFR. I learned the type of capacitor makes a bigger difference to listening to music than a 50% change in the value. My final step is to get some of the 100pF Evox PFR caps to compare to the stock 220pF. And I won't bother comparing with the 100pF polystyrene.

This was a stimulating and enjoyable experience, for which I thank you Graham.

Posted By: Fatmangolf
Date Posted: 26 Aug 2010 at 7:27pm

Noting again my thanks for Graham's steer on obtaining Evox EFR, here is an update.

The Evox EFR 100pF capacitors have now been in the Genera for a day or so. I left them on in parallel with the 220pF's. There is an audible difference between 100pF and 220pF, so I am glad I tried this. It is quite a small and subtle impact on the high frequencies (like a tone control as Graham has pointed out elsewhere) and could easily be masked by a less transparent preamp than the Genera, which is a great design with carefully selected parts.

Posted By: Graham Slee
Date Posted: 31 Aug 2010 at 12:14am

I prefer to model a cartridge output pre-equalised.

A magnetic cartridge output is a rising response at +6dB per octave.

I developed a cartridge model to do this...

The voltage controlled voltage source (VCVS) provides a zero impedance signal source to build the cartridge model to. The input circuit of the VCVS provides the rising response. Probe 2 measures the rising response. The 570mH inductor, 660 Ohm resistor and 1pf interwinding capacitance model the Goldring 1042. The interwinding capacitance isn't quoted by Goldring but even so there will be some - I chose an arbitrary 1pf. The 75pf cap is the arm capacitance I use. The rest is the input load of the Genera. L2 models the input DC blocking cap inductance and values of 30nH (a typical good electrolytic) or thereabouts don't show until " much closer to the frequency of light"

The 220pf cap we've been talking about is depicted in red. All values can be played with and the resulting charts are as follows...

This one shows the response to 1 gig! The reason I do it is to see what the amp input sees. Most audio op-amps have a Gain Band Width (GBW) product of 4 - 10MHz and it's important to consider what's going in to their inputs in the form of circuit stimulus whether real or imaginary - the simulator sees them as real.

The peaky response (up nearly 1.4dB from the curve at about 8kHz) is the 220pf load cap. The lesser one is 100pf. The red line is the rising response.

You can probably see why I chose 220pf by examining the stimulus at 4-10MHz. When the interwinding capacitance of the cartridge is factored in, the curve which otherwise would be well down at 4-10MHz is kicking back up due to this other LC response. With 220pf it's 6dB down on the 100pf curve. This suggests the input slew rate is half of what it is with 100pf and although stating the obvious, it shows how these curves help indicate what's going on. If the input stimulus is trying to race its output it could muddy the soundstage. The trade-off being the slight treble peak. The audibility of the peak is going to vary with the listener's age and many other things.

Reducing the load capacitance further than 100pf would only exacerbate the slew rate "hunting", plus the bottom end of the available polypropylene capacitor range is 100pf. We've already heard that the PS caps don't have the same sound. Going higher than 220pf just emphasizes the peak further.

Therefore, from a capacitive standpoint it seems the choice is 100pf or 220pf. We could do a series parallel arrangement to obtain 150pf I guess, but just look at the additional "material" involved. Yes, I think it would be a case of splitting hairs.

The only other place to go could be resistive loading. I'll take a look at that next.

-------------
That none should be able to park up and enjoy the view without a smartphone and the knowledge in how to use apps

Posted By: Fatmangolf
Date Posted: 31 Aug 2010 at 10:34am

Graham, thank you for sharing that model and your explanation with us. It puts my Spice model to shame and has given new insight (to me at least) into the behaviour of the cartridge/preamp circuit.

Yes I ordered enough of the Evox PFR 100pF capacitors to try them in series for 50pF or series parallel for 150pF. I haven't tried either and the latter is quite a small proportionate change, giving combined capacitance of 250pF in the in the middle of the 200pF and 320pf options already available. It would be "Splitting hairs" indeed. And a squeeze to get the parts onto the PCB.

The other variable is the arm capacitance, C4 in Graham's model. There's a limit to how low this will go with available cables, practical TT to preamp distance and enough RF shielding. My internal arm and external leads measured at 105pF, lets say 100pF. +/- 10%. I like the internal Litz wire I fitted but changing the external lead I used may (!) get the arm capacitance down to 50 or 60pF with a short, low capacitance cable. I would prefer that to further reducing C2 inside the preamp. Usual warnings to all about voiding warranty, being competent to do the work, and the point that different cable compositions do sound different.

To be honest I am reluctant to rewire as I love the sound of my TT and cartridge at the moment. I would credit the Genera with the clear and engaging sound I am hearing. I tried another well regarded preamp which I have souped up with better parts and the Genera knocks it for six on every ball.

Back to topic, I would be very interested in the effect of resistive loading in this model. With another preamp I have tried lower loads i.e. resistor more than 50k, and found they sounded a bit thin and less clear at HF, the word "phasey" catches it. Has anyone else explored this?

Jon

Posted By: Fatmangolf
Date Posted: 09 Oct 2010 at 9:13pm

A follow on to the Evox PFR "little gems" note above.

Whilst looking inside my Arcam Alpha 8SE earlier (which still sounds very good 10+ years on, BTW), I noted the component choices in the DAC module. Lots of Rubycons and a lot of little blue capacitors, yes they were Evox PFR!

It helps to explain my continuing use of the Alpha 8SE (when, ahem, I'm not listening even more keenly to my Rega and GSP Genera, of course) and suggests great hi-fi minds think (and ears hear) alike.

Posted By: morris_minor
Date Posted: 08 Dec 2012 at 12:07pm

Bit of a thread resurrection here!

I make no pretence of understanding the technicalities of cartridge loading, and I'm a strictly plug'n'play kind of person, so any kind of circuit component changes are definitely not on the cards.

So my question is: seeing as how cartridge loading can change the sound, how can this be achieve simply?

There is an additional complication as I have two Technics DD parallel trackers (SL7 and a hot-rodded SL-QL1). These take p-mount cartridges that just plug into the end of the tone arm, so there are no headshell leads (which is where, I guess, some changes could be made).

I've nothing "wrong" with my vinyl sound, but I guess like most of us here, am always looking out for some incremental improvements, and this seems like an area to look at as I've read (dangerous! Embarrassed

) that no two cartridges need exactly the same loading.

So I'm hoping for some enlightenment!

-------------
Bob

Majestic DAC/pre-amp
Accession MC/Enigma, Accession MM, Reflex M, Elevator EXP, Era Gold V
Solo ULDE, Novo, Lautus USB and digital, Libran balanced, CuSat50
2 x Proprius + Spatia/Spatia Links

Posted By: tg [RIP]
Date Posted: 08 Dec 2012 at 12:19pm

http://audio-forum.gspaudio.co.uk/topic1246.html - This could be what you are looking for Bob Smile - at least the "how to" part.

Posted By: morris_minor
Date Posted: 08 Dec 2012 at 2:59pm

Thanks tg! That's excellent Clap

I just need to work out what values to choose now. Oh, and buy a cheap soldering iron. SWMBO will have the fire extinguisher poised if she sees it smoking . . LOL

Posted By: Fatmangolf
Date Posted: 09 Dec 2012 at 5:00pm

I had forgotten this thread but not Tony's helpful one about adjusting the loading with plugs. Thank you! Great examples of the range and depth of advice we share.

Jon

Posted By: morris_minor
Date Posted: 09 Dec 2012 at 5:44pm

You're right Jon. I was just going to start a new thread when I thought I'd do a little digging and unearthed this gem ...