CONUNDRUM

Elevick, the PH-7, if modified by ARC, has 6 tubes: (4) 6922 tubes in the gain circuit; and (1) 6H30 tube and (1) 5881 tube in the PS. I don't think your idea has legs.

Handymann, the Koetsu is a very expensive cartridge, most models are in excess of $4K. Not sure, but there appears to be several Urushi models. Based on a very high-level check on the Needle Doctor web site, it appears the output voltages range between .2mV and .6mV. This class of cartridge appears to have a low compliance factor of 5.

Clearly, the Koetsu generates much less voltage than the Maestro. Therefore the PH-7's output should be less too.

If you want to try a LOMC, perhaps it would be prudent to try a cheap LOMC and see how it works. As I mentioned above, the PH-7 has a noise rating of 70db. It would be very unfortunate if the PH-7's output put out alot of noise with low output carty. I reiterate, it's worth a 15 min call to Calvin at ARC. He's in the best position to advise you.

I'm not 100% sure this will help, but here are a couple of things to consider:

From the Calypso manual : Note: The input circuitry of the Calypso accepts a maximum input voltage of 3.5v rms single-ended or 7v rms balanced. This is well above the industry standard 2v rms single- ended or 4v rms balanced. However, there are a few source components that exceed this limit. If the offending source component has a volume or gain control, all that is necessary is to reduce the gain or volume of the source component. In other cases, it will be necessary to have the source component modified to be within industry standard specifications. A clear indication that this limit is being exceeded is a raspy distortion or compression of musical peaks.

From another Audiogoner, on changing the gain of the Calypso:

To change the gain on the Calypso, turn power off for 20 mins, remove top, and locate jumper block for each channel. To find this, locate V2 tube (the left hand one), move right past a block of four resistors, and you should find the jumper block (just to left of two yellow capacitors). The jumper block has four positions, and if the Calypso is on high gain, there should be a shunt in the upper one and the lower one, with two empty positions in between. To change to low gain, move both shunts from the outside to the inside positions. Then repeat for the other channel.

Note: older versions (at least of the Janus) did not have this jumper block in which case obviously you can't adjust the gain. If in any doubt, email glenn@aesthetix.net and he can send you detailed instructions/diagrams,

The ARC phono stage with your cartridge is outputting 2.7 volts to the line stage. This is more than enough to drive your amps w/o the linestage. Adding the gain of the linestage, you now have a whopping 38 volts of output to drive the amps at full tilt. You need to either change out the cartridge to match the phono stage or change the phono stage the match the cartridge. I wouldn't mess around trying to mod the phono stage - who knows how that will sound.

Two suggestions. Clearaudio has a reasonable trade up program. Look for a model that has output less than 1.0mv. That should take care of the problem. On the other side of the coin, I would consider looking at a Aesthetix Rhea. It will work well woth your line stage and you get variable gain with the ability to voice it as you like by tube selection.

The input impedance of the Calypso is 20kohm if unbalanced, 40kohm if balanced.

The output impedance of the PH-7 is 200 ohms unbalanced. According to ARC, the output load should be a minimum of 10kohm, so let's assume that value as the total load, and estimate the values of some resistor networks that could be built into the interconnect between the PH-7 and Aesthetix.

If you want to reduce the PH-7's output to half (-6dB), you would need two 4.99kohm resistors per signal polarity. One resistor pair if you use the unbalanced outputs from the PH-7, two resistor pairs if you use the balanced outputs.

One of the 4.99k resistors should be soldered in series with the "hot" signal wire of the interconnect cable, the other resistor is soldered from the output side of the first 4.99kohm resistor to ground (outer shell of the RCA connector).

However, we must keep in mind that the Calypso already has a 20kohm input resistor, so for the second resistor (from output to ground), we want a value that forms 4.99kohm when placed in parallel with the Calypso's 20kohm resistor.

One such value for the second resistor would be 6.8kohm, which in parallel with 20kohm, will form 5075 ohms, which is a +1.7% error.

Another possible value for the second resistor would be 6.65kohm, which in parallel with 20kohm, forms 4.991kohm, for an error of +0.01%.

In either case, the first resistor remains 4.99kohm.

For a reduction to one-quarter (-12dB), the series resistor should be 7.5kohm and the ground-side resistance should be 2.49kohm.

Again, since the Calypso already has 20kohm at its inputs, we want a value for the second resistor that forms 2.49kohm when placed in parallel with the Calypso's 20kohm resistor.

One possible value is 2.7kohm, which in parallel with 20kohm, forms 2.379kohm, which is an error of -4.46%.

A more suitable value would be 2.87kohm, which in parallel with 20kohm, forms 2.51kohm, which is an error of +0.8%.

If your ears suggest that the total 10kohm load is too heavy for the PH-7 (slightly muted dynamics, slight loss in resolution), generally you would double all of the resistance values. 10kohm becomes 20kohm, 4.99kohm becomes 10 kohms, 7.5kohm becomes 15kohms, 2.49kohm becomes 5kohm. But all of the second resistors will need to be re-calculated, I suggest that you try the 10kohm setting first, and if that isn't to your liking, let us know and I or someone else will do the math for 20kohm instead.

Please keep in mind that the lower resistance, the lower the noise, so I would not increase the resistor values unless I thought that the sound lacked life.

Theoretically, the resistor network could be placed either at the output of the PH-7, or the input of the Calypso, but if you place it at the output of the PH-7, you will get a roll-off in the high frequencies, due to the capacitance of the interconnect. Better to place the resistor network at the input of the Calypso.

I believe that the above will be the easiest and most economical solution to your troubles.

kind regards, jonathan carr

FWIW, the Rothwell Attenuators, and I would imagine most other comparable devices, are configured in the same manner as what Jonathan has described, with a series resistor, and with a shunt resistor at their output. I have a pair of the Rothwell's, and as measured with my not particularly accurate analog multimeter the series resistor has a value of about 21K, and the shunt resistor has a value of about 9.5K. In conjunction with the Calypso's 20K unbalanced input impedance (the PH-7 does not provide balanced outputs), that will result in an attenuation of about 12.6db. The load impedance seen by the PH-7 with the attenuators in place would be a presumably comfortable 27.4K. The attenuators are intended to be placed directly at the input connectors of the destination device (the Calypso in this case), so cable capacitance would not be an issue.

Handymann, while I'm not sure that 12.6db will be a sufficient amount of attenuation (given that the cartridge will probably exceed its 3.6mv nominal output by a considerable amount on the peaks of some material), I repeat my earlier comment that I'm surprised that you reported significant sonic degradation using an attenuator. Can you indicate what model attenuator you used, how much attenuation it provided, and exactly where in the system it was connected?

Regards,
-- Al