Moving Coil Preamplifier

I'd have thought it depends what you are amplifying. What frequency range, amplitude of output etc and for what purpose?
You can proabably knock up something acceptable with a dirt cheap single rail quad opamp, AC coupled as long as you don't want Hi Fi.

Del

If your output is 2mV, then this is a 'high-output' MC cartridge, and you can use an ordinary MM preamp ('low-output' MC cartridges are less than 1mV, usually around 0.3mV, and require an extra gain stage in the preamp, or a step up transformer).

The main consideration is good quality components. Outside of this, the subject is too deep and involved to cover here, but there is a lot of interesting information available on the web via Google, and there are plenty of kits and schematics to be found.

There have been some pretty good home brew MC pre-amps on the web since the web began. Do you want to include RIAA EQ or not?

I have seen some audio op amps with some really good numbers lately, if you are adventurous you could try to roll your own. Keeping noise low is the largest problem. I have seen several battery operated preamps that eliminated ground loops and power supply hum that way.

Don't forget what the people from Monster Cable say, if you don't use grain aligned Oxigen free 00 guage wire for your cables you are waisting your time, it'll never work.

Hi Tommytech,

No need for RIAA. I have build few preamps using opamps but my latest use only fets and I like it better.

Do you think fets would be a good choice for such a low voltage/impedance? Or it is only a matter of circuit design? Most of what I found on the web use transistors, sometime four in parallel...

Thanks

"No need for RIAA". I don't get your meaning here. Unless you want to lose all your bass I would have thought RIAA is critical.

Hi Holzfeller,

the RIAA equalization is already a part of the preamp with fets I have mentioned.

All I need is a linear pre-preamp.

Understood. I am a huge fan of keeping things as simple as possible, so a single extra gain stage should be enough. I have not built or heard the BHL phono preamp, but I like the look of it. This rendition also includes a nice and simple pre-preamp. I am also a fan of jfets, and I am building onboard preamps for three of my guitars using mostly jfets. I considered building a phono preamp, but already have too many unstarted projects (and many boxes of components), but maybe someday. Here are some others I also like the look of.

I think I will build the Le Pacific MC PrePre .I'll let you know how it went.

Debating various fets, transistors, tubes with real aficionado, with golden ears, is, ahem rewardless exercise. They can hear, normal mortals cannot, and can distinguish things, even more than mortals can.

I do not know, which is which. And life is too short.

Other than that, the presented diagram is a promising one.

Great, let us know how it goes.

But I would still question whether this is necessary in the first place. The MC PrePre is designed for boosting the output of a 'Low Output' MC cartridge (typically 0.3mV) to around the output of a MM cartridge (typically around 4mV). If you use this with your 'High Output' 2mV MC cartridge it is very likely you will overload the preamp. Overdriven jfets might sound quite good in a guitar preamp, but are not wanted in a hifi preamp, not to mention the damage that could occur.

On the other hand, you might want to use this circuit to make adjustments to input impedance and capacitance (the former is typically 47k Ohm for MM and 100 Ohm for MC. I think R8 sets the input impedance in this circuit. My phono preamp also has selectable input capacitance, 100pF/200pF/320pF.), and reduce the gain to close to zero (change the ratio of R9 to R11).

Didn't know if your preamp had a phono input. Started becoming rare years ago during that null in vinyl.

You can use a op amp. One should be OK for a standard 100 mV output. Just go for low noise types. (in my amp days they had a suffix c) It all depends to what voltage you want to crank it up.

For signal in the range of 2mV, thermal noise becomes a problem as transistor base emitter operating voltage drifts at about 2mV/oC. Low temperature drift differential amplifier having 10Hz to 100kHz working BW, high common mode rejection and gain of X1000 will be ideal.

Impedance booster matching transformer can also increase voltage X100 level and then it is easier to amplify the signal. JFET offer lower noise down to 0.3nV/vHz and 20Hz to 20kHz range this figure may be much better for 1kHz to 20kHz frequency zone.

For JFET visit www.interfet.com

It's actually the lay out and construction that is easilly as important as the circuit design.

Not to mention the power supply.
Del

that makes sense, Del the cat,

I would like to avoid batteries which (I assume) would deliver the best DC.

So, in a power supply (with a 78XX) what kind of capacitance should I use? Or there are other factors to take into consideration? Are you thinking about ground loops?

Thanks!!!

78xxx data sheet will tell you waht you need but generally 0.01uF disc ceramics from input to gnd and output to gnd will give hf stability 10uF or so for general smoothing.
I'm assuming you are doing this for fun rather than necessity, in which case the LM324 is a dirt cheap handy single rail OPamp for experimenting with (I've actually use one in a military headset mic amp before now shhhhhh)

Ground layout... everyone has their pet way of doing it, but I'd suggest leaving points where you can break or link gnd tracks to allow for experiment.

I remember the first Hi Fi preamp I built, it worked fine but would suddenly go int motorboating or pick up local taxi radio! (I cut the gain down a tad and it was ok)

Del

Not necessary. Coil has very low impedance and current through coil fairly large enough does not require any special consideration. Wrong circuit should not be used. One has to use low input impedance circuit and not high input impedance circuit is the only good choice.

One very important thing to consider is the impedance the amplifier is looking for. From the output of your pre-amp, you need to make sure the impedance is within specs of what will work with your amplifier.

Don't forget to also check the maximum output you can feed into the amplifier.
Again, you can cause damage to the amplifier or have a distorted sound. Too low output = s/n problems. Too high output = distortion and if too high, a blown amplifier.

For audio generally 600 ohms impedance matching transformers are used. Balanced input with very high common mode rejection is ideal to get rid of AC humming noise. Keeping frequency band between 400Hz to 4kHz does some good for human voice but for true audio range 20Hz to 20kHz entire frequency response need to be of flat response.

If you use 78xx for powering the preamp, it will not sound good. You have to use a much better psu like a shunt regulator.

For a PrePre, a better power supply is a must.

Gajanan Phadte

Could you not use the PSU from the existing preamp?

Otherwise, for such a small circuit with low current draw, I would be inclined to use batteries. This would give good results and avoid the usual PSU problems.

But again, to avoid any additional noise or losses, I wouldn't bother with a PrePre when you already have high output 2mV MC cartridge.

I totally agree that at 2mV a pre-preamp is most likely not required. A friend with 35yrs repairing, modifying and building ultra high-end gear did direct 2mV MC connections to MM preamps twice with complete satisfaction from demanding customers. Yet, there are two cases for which I can see some use for the pre-pre at 2mV.

1. The MM preamp is already almost too noisy for an MM and would be about 6dB worst with a 2mV MC. Note however that preamps with a low Z feedback network and a BJT transistor base connected directly to their input with little or no series resistance might show reduced noise when the MC low impedance shorts their input noise current. So even with noisy MM performance, it is worth trying direct connection since the noise reduction might outweight the signal reduction.

2. If you have multiple sources connected to the same volume control, you may be anoyed by the loudness boost when switching to a high output device like CD player after setting the volume for listening to a record, and even worst a quiet one !

Note that my friend added load resistors matched to the specific MC cartridge, which he installed inside the MM preamps directly on input connectors. This brings me to a reply to my previous post #14 that needs to be put in perspective. My post was about an input resistor free common base BJT design.

Quote: "Remember that pickup coil has dynamic impedance for frequency due to its inductance and not just its DC resistance. ZL = jwL "

MC cartridges have very low inductance and provide flat response when resitively loaded, which is the load type this common base amp provides. Loading resistance is adjusted by the DC bias current. Proper loading resistance is usually about 10 times the resistance of the coil but this ratio vary widely from model to model. You should verify the value recommended by the manufacturer for your model and see what experienced audiophiles think about it. No load value provides perfect flatness so it becomes a matter of taste and global audio chain balance.

MC are relatively insensitive to small capacitive loads such as cables. Values near 0.1uF are needed to noticeably affect the freq response. Capacitive loading causes a peak in high freq response, normally to be avoided, but can also be used in some cases to balance high freq roll-off of the MC or also any part of the audio chain. The following article explains well and short http://www.extremephono.com/Loading.htm . This other, more detailed with calculators http://www.hagtech.com/loading.html (In this last I only disagree with the capacitance always bad conclusion.)

Quote: "Voltage gain = Collector resistance / emitter resistance (that of coil)"

This formula assumes the external resistance to be much larger than the intrinsic emitter resistance. In this case, the bias is low enough such that the intrinsic emitter dynamic resistance (1/(Gm-Gm/hfe))@Ic equals the desired load impedance. This impedance also defines the gain since the virtual ground is at the base rather than the emitter. The coil resistance does not need to be accounted for since it already was when the cartridge output voltage was measured at the specified load. It would also be the same for an highZ preamp with the right load resistance. Also, the current mirrors that send the signal back to the grounded load can have a current gain up to about 30 with good linearity and gain tolerance, and in fact 900 with darlingtons. So mirror gain must be factored into the formula which becomes:

Vgain = (Gm-Gm/hfe) * MirrorGain * OutputLoadResistance

In Kokkoplus case, a gain of only 2 to 4 is needed to drive an MM preamp so for 200 Ohms input Z (20 Ohms x 10) the mirrors only need a gain of 1 with a 400 to 800 Ohms output resistor.

This said, I found an issue with this input resistor free common base design that brings a new consideration for BJT designs:

The noise level of a BJT varies with bias current. The noise vs current graph curve is U shaped. This means there is an optimal operating current for lowest noise for a given BJT model. If this current happens to provide the appropriate input impedance for the input resistor free design, nobody would know about Murphy's Law ! However, since the bottom of the curve has some width and there is more than one BJT model to choose from, not all hopes are gone. My repair friend told me that at least 2 high-end MC pre-amps use common base, but he could not remember if they were input resistor free. I will do some more investigation and then simulations to satisfy my curiosity and will post my findings later.

Another balancing act has to be considered for all BJT designs. As Shyam mentioned previously, noise increases with temperature. So the higher the bias, the greater the temp rise and so the noise. The higher the Vce voltage the worst the effect, the more heatsinking the least. So for a given Vce and Heatsink, the optimal bias will be more or less to the left of the lowest point on the datasheet curve, but in all cases to the left. Base, emitter and collector resistances also interact with bias but likely in both directions since values chosen tend to go down with bias increase.

Now to Kokkoplus, you first have to try direct connection to your MM preamp (without loads is fine) to check if the noise level and output level are satisfactory. If yes, all you have left to do is find the load recommended for your cartridge and get the audio grade resistors. This is unless you are craving for a pre-pre pet project.