Eros phono RIAA low frequency performance

[milosz]

I'm seeing more low frequency rolloff with the Eros than I expected- around 9 dB at 20 Hz.  Is this typical?

The measurement was made using the Hagerman inverse RIAA network. 

Note:  Other phono stages ( Parasound Z-Phono, Conrad-Johnson PV10a preamp phono stage)  that I've measured show much less rollof,  about 1 dB at 20 Hz, so I don't think this observation of a rolloff with the Eros is an artifact of my measurement method.

Bottlehead Eros
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Flf.org%2Fmilosz%2Ftemp%2FBhE-fr.png&hash=a642ed41b806f92bc68b0d160345dff1f9aa6115)

Conrad-Johnson PV10a phono stage (for comparison)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Flf.org%2Fmilosz%2Ftemp%2FCJ10-fr.png&hash=0178f4dac40dcee415f5d570ef012d5421c59879)

[Doc B.]

It's not typical. You might want to check the connections of the caps in the RIAA network.

[Paul Birkeland]

Are you using the IRIAA at the input or the output of the Eros?

[grufti]

Your result is not the norm. I used the Hagerman inverse RIAA to measure my Eros when I built it many years ago. It was about -1dB at 20Hz in stock form.

[Paul Joppa]

If the RIAA caps are solid - and the right values - then I would suggest measuring the 6922 voltages, in particular the bias voltages C2-C3 and C7-C8. If the tube is failing, the bias will be low and the resulting bit of grid current will drag down the impedance, blowing the LF equalization out of the water.

Amusing sidelight - this problem occurred in the first iteration of this circuit topology, and took a long time to sort out. By which time we had decided the circuit was junk. It languished for another five years before we tried it again in the Tape Head Preamp, which begat the Eros.

The only "extra" LF cutoff (more than RIAA) in the Eros is the output capacitor, which is -3dB at 1.6Hz into a 100K load.

[milosz]

I am using the inverse RIAA network on the INPUT of the preamp.  I have calibrated my test signal level to the phono preamp to be 5 mV  at 1000 Hz (measured at the input jack of the preamp)

I am not using the stock 6922 tube, I have a "New Old Stock"  United 6922.

I'll try the stock tube.  Maybe this "NOS"  tube wasn't really that "N".....!

I'll also check the RIAA caps and resistors.

FYI  I built this Eros some time ago, and only listened to it briefly to verify it worked, didn't hum, etc.  I've been fooling around with audio measurements and decided to measure the various phono preamps I have laying around...  I plan on adding phono to my system but need a proper table or shelf.   Right now my VPI Scout in on a table in my workshop with my collection of phono preamps and a headphone amplifer.

[milosz]

OK, so I put the Electro Harmonix 6922 tube instead of that "NOS" tube into the Eros and now the frequency response looks like:

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Flf.org%2Fmilosz%2Ftemp%2Feros.png&hash=e5e1096ff5e3674373de372c473b27e458d79d7d)

...which probably means "perfectly flat to the limits of my measurement system's ability to measure"

In this case "NOS" 6922 means "Nasty Old Stock"  I guess!

Thanks for mentioning that a weak tube can cause a low-frequency rolloff.  I would have thought that a weak tube would just have low output, higher distortion, not a frequency-dependent character.

The Electro Harmonix tubes are actually pretty good, probably about the best "obtainable, in-production" tubes that are out there.

I have ordered a new "Winged C"  6J32P to try - is this a drop-in replacement or do I need to rewire the socket & etc?

[Paul Birkeland]

Thanks for following up on the measurements!

The 6J32P will pop in and work quite well.