Preamp for EPC-450CII and semiconductor / strain-gauge carts

[williamsunique]

Oh yes I forgot to say out voltages are now minus .79v and minus 2.52v.
Paul

[flavio81]

Williamsunique, what is HT?

Flavio, I don't see any harm in using your super-simple as an intro for those having the carts until you get the new one sorted out. BTW forgive my presumption but I am concerned about the new front end vs the super-simple in that the SS cannot ruin a cart by overcurrent thru malfunction while I'm not convinced the new circuit has this safety margin.

Why should there be a difference? Both are using the SG as a voltage divider in the same way. Or you mean because of the 1K "mixing" resistors and the bias compensation voltage?

Current limiting of resistors is pretty stable and reliable, active devices can go wrong. I wouldn't use a current source to bias these carts.

There is a misunderstanding here, and it's caused by the way i have drawn the 1st stage schematic.

I am NOT using a current source. The cartridges are biased the same way as in the "simple preamp": By connecting them to a +V (9V) voltage source, through a circa 2.2K resistor.

The voltage that appears on the circuit and that i choose to call "bias" is NOT biasing the strain gauges; it is biasing the output voltage so the DC offset is removed.

Anyways, it seems that, to make you guys happy, i should redesign the 1st stage to a more "conventional" way, in other words, to an similar way to my "ultra simple" preamp. I want to make you happy so i will do it.

But the idea with my "bias" circuit was to avoid having to use DC removing capacitors of high values... "For great justice".

[SLA-3]

I know the original circuit works fine without any unintended consequences 'got-chas', I just want to be sure the new circuit has the same safety margin. I do not see what the operating voltages of the front end stage is.
Also, what is the design loading of the output of the gain stage end? Minimum/maximum loading?

[williamsunique]

Flavio, sorry poor terminology. HT high tension. Should have said "Power rails" plus and minus 9volt.
Paul.

[williamsunique]

Flavio, yes far better to bias the SG through 2.2k from a 9volt battery, rather than a 3volt battery connected straight across the SG. If the power rails are plus and minus 9 volt as well, then under fault conditions the worst case scenario (unlikely) would be circa 7 to 8 mA flowing through the SG.
Paul.

[williamsunique]

Hazard wrote. "The Verdict: This is the best cartridge I have ever listened to. Even with flat gain, no equalisation. The treble is FAST."
Maybe your cartridge does not require some adjustment to eq. The EPC-450's are certainly better with.
Paul

[williamsunique]

I have had a chance to apply a variable voltage source to the negative input of X3. The minimum outputs obtained on X1 and X2 were 0 volts and 1.4 volts. The input voltage at X3 with the outputs as close to zero as possible was 1.25 volts. This was with a power source of plus and minus 9 volts to the opamps. I connected 2 x 9 volt batteries with 2.2K ohm resistors in series with the SG elements (820R resistors as simulation). Both strain gauge elements polarized the same way. The current flowing through the SG (820R) was 2 mA. I then substituted the 2.2K series resistors for 1K and now the current flow was 3.5mA.
I disconnected the bias to the SG elements and fed a line source (music) to the preamp inputs. The preamp outputs were connected to an amplifier and speakers. There was no output however while bias was applied to X3 negative input. I was expecting output, should I have had?
Paul

[williamsunique]

I said. "There was no output however while bias was applied to X3 negative input." This is not strictly true. I turned the amp up full and there was very low distorted sound.
Paul.

[flavio81]

Flavio, yes far better to bias the SG through 2.2k from a 9volt battery, rather than a 3volt battery connected straight across the SG.

I never NEVER EVER EVER, NEVER, NEVER, NEVER said or suggested this!!

I should have never posted that schematic. Oh dear...
Anyway, i will post a new schematic once i get enough time to draw it.

[hazard]

Flavio, yes far better to bias the SG through 2.2k from a 9volt battery, rather than a 3volt battery connected straight across the SG. If the power rails are plus and minus 9 volt as well, then under fault conditions the worst case scenario (unlikely) would be circa 7 to 8 mA flowing through the SG.
Paul.

Well yes it is far better to to bias the SG through a 2.2k resitior - because it won't work at all connected straight across a 3 volt battery. Forget the nonesense about current through the SG under worst case scenario, this is not the point. If the strain gauge is powered directly from a bettery then the voltage across the strain gauge will not vary with resistance change, there is no "output" from the SG. I use a current source to bias the SG, better than a resistor, but as previoulsy poined out if resistance changes are small then a 2.2k resistor is close to a current source.

[SLA-3]

While I do admire the ingenuity of the 'new' front end ckt, it seems to be a 'fix' for a non-existent problem re the cart's polarity and grounding. There is no potential grounding problem I can find with them to be of concern, and effective element polarity seems to be a non-issue outside of establishing correct acoustic polarity output.
I had some extended conversations with someone very familiar with SG carts and their support ckts and he said that there was no audible benefit to a current source over simple loading resistors. He also didn't think there was any sort of "polarity" to the SG elements themselves IIRC.
If there is a bi-polar voltage source available (the Pana-Tech CD-4 SG adapters have mono-polar PS) I see every reason to keep it simple with your elegant original ckt taking advantage of such with the basic and effective foolproof loading resistor network rather than the 3 IC approach. Have this original high-gain ckt coupled into and driving a passive correction network followed by a line driver buffer stage with enuff gain to make up for the insertion loss of the correction network.

[flavio81]

While I do admire the ingenuity of the 'new' front end ckt, it seems to be a 'fix' for a non-existent problem re the cart's polarity and grounding. There is no potential grounding problem I can find with them to be of concern, and effective element polarity seems to be a non-issue outside of establishing correct acoustic polarity output.
I had some extended conversations with someone very familiar with SG carts and their support ckts and he said that there was no audible benefit to a current source over simple loading resistors. He also didn't think there was any sort of "polarity" to the SG elements themselves IIRC.
If there is a bi-polar voltage source available (the Pana-Tech CD-4 SG adapters have mono-polar PS) I see every reason to keep it simple with your elegant original ckt taking advantage of such with the basic and effective foolproof loading resistor network rather than the 3 IC approach. Have this original high-gain ckt coupled into and driving a passive correction network followed by a line driver buffer stage with enuff gain to make up for the insertion loss of the correction network.

I am taking note!! Thanks!!

[flavio81]

IMPORTANT UPDATE

So i installed again the EPC-450CII on one of my turntables using the "ultra simple" preamp, but this time using a 31-band equalizer to dial in a preferred EQ.

Then i A/B compare the same record's sound using another turntable which is fitted with a Grado MCX low output cartridge (fitted with a Grado Red1 stylus) which IMHO has a very nice, appealing tonal balance. So this Grado is my "reference" for tonal balance. Plus it's sounding nice, nicer than i expected from a Grado.

Comparing both, i see that my EQ correction circuit (posted before) will not be so good. The "valley" reported in tests between 2-6KHz is IMHO necessary for perceiving the sound as "smooth". Such a "valley" appears in cartridges like the Grado Prestige line (i.e. Red) and the M97xE and i'm totally sure this is what gives them a "smooth" sound quality. My EQ correction circuit would flatten said valley, and now i'm convincing myself that this isn't so desirable.

On the other hand, the 250-650Hz range definitely benefits from a 2.5-3dB reduction. Deep bass could be enhanced too.

Finally, the 15-20KHz frequencies are way too loud with this cartridge. I think a low pass filter with cutoff frequency between 10 and 14KHz definitely helps a lot with the cymbals. I'm currently trying 12KHz with great results.

All in all, it seems i'll have to re-design the EQ circuit. Don't hate me

Good news is that this cartridge indeed is superior sounding...

[dlaloum]

Heh it seems to me that you are saying that when EQ'd for "flat" frequency response, it starts to sound like CD....

Most high end audio room setup guys "tune" the rooms (and speakers, etc..) for a response that gently rolls off the high end in a Natural way...

In my opinion the perfect setup would do the "pleasing sound" tuning at the room level, and not the source level. - At the source it should be perfectly Neutral... That way all your sources can be flat, and they all end up sounding good.

Also natural acoustic auditoria (those considered to be the best by classical music performers) also tend to have a similar type of rolled off top end profile...

bye for now

David