yep absolutely true.
I do think that for low impedance MC cart, the BJT is more suitable than the JFet.
It seems to be a non sense since pass has been using the Jfet for the MC input and I am sure it works great, but the nosie floor of a BJT for low impedance input is definitely betetr than the JFETs thus make it IMHO more suitale.
also trye. However to have a fair comparison, the RIAA shall remain the same.
You know my fetish for V-Caps
This is a heck of a frequency response.
Bass response cuts off at 0.4Hz,while upper end at 34MHz.
The low end might pose some problem is turntable or set up is not rumble free.
For the kick these are also the values for the RIAA.
With these simulated values, precision is roughly abs(0.1dB) within the audio band.
Enjoy.
Bass response cuts off at 0.4Hz,while upper end at 34MHz.
The low end might pose some problem is turntable or set up is not rumble free.
For the kick these are also the values for the RIAA.
With these simulated values, precision is roughly abs(0.1dB) within the audio band.
Enjoy.
Attachments
Without FET's or DC-servo, remaining option is to trim the input stage periodically.
There I go again, if you where using the 'servo-that-is-keeping-dc-differential-away-from-the-input' then you could 'tune' it to create a filter removing 5Hz and below.
I still don't think that the input DC will drift so significantly to require periodic trimming or the use of servo.
Especially if monolithyc NPN and PNP are used.
Maybe I am wrong, but on the built I have I don't see any drifting on the input voltage.
Maybe I should check that better.
I wasn't saying that in a negative tone, I was saying that the overall design extends to very deep low end.
You can also remove this on the servos that control output offset or tune it on your RIAA filter or adding it at the output. There are several options.
I still think that the servo injecting at the differential input is intrusive, I mean servoing right at the input seems a bit excessive to me and if it can be avoid I would try to do so.
You can build it with and try to listen with and withouth to see if it makes any subjective difference and perhaps post your impression and decide whether to keep it or not afterall.
I would push the evnvelope even further, since we are doing all discrete, I will try discrete servos using K389 and a simple op amp circuit.
I understand many people will be against this, and not sure if this will yeald to a sonical improvement for sure or not.
However, quick experiment I did in the past was to replace the servo on my previous phono design with the descrete DEXA opamp.
Sonic improvement was significant.
Anybody here have tried to do the same thing and wants to share their experience?
Strangely enough, others do seem to think so, and trimpots are cheap.
Me think, what's the use of a 5000 buck cartridge if it also ends up getting all-mesmerised, and magnetised ?
(e.g. a 0.3-0.35mV DV-XV1)
Oh, look at the time already, have to go sailing with Mr Børresen, before he's off fishing in some obscure Scottish waters.
Last edited:
All fine with me.
But now you are loosing me.
It just creates more developer's pain, and no gain
This I have great trouble believing
especially when using servo-output-filters like as I do.Not me
Last edited:
Quick question for Frans, the master of the servo design (By the way Thanks again for helping out with this!).
I just got this idea, but I am not sure it can be put in practice. Would it be possible to replace the two single OPAmps with the Differential OPA1632 somehow?
That would probably get the best of the worlds.
It is probably a silly idea if so my apologize for that
I just got this idea, but I am not sure it can be put in practice. Would it be possible to replace the two single OPAmps with the Differential OPA1632 somehow?
That would probably get the best of the worlds.
It is probably a silly idea if so my apologize for that
Do you have a XV1? Wonderful Cartridge!
My question is how much input offset do you consider too much? If I put my DVM across input I measre 0V. That is open input though, I don't put meter across cartridge to not risk to burn coils (hilarious because here we are talking about offset injected).
Should I perform a different measurement?
Perhaps connecting the equivalemnt of the internal load of my cartridge across the input and put the meter?
Since my cart has 5ohm impedance I figure input offset will be even lower than with the 200ohm input impedance set as a default.
Maybe I am wrong on this.
Quick question for Frans, the master of the servo design (By the way Thanks again for helping out with this!).
I just got this idea, but I am not sure it can be put in practice. Would it be possible to replace the two single OPAmps with the Differential OPA1632 somehow?
That would probably get the best of the worlds.
It is probably a silly idea if so my apologize for that
If not silly
then at the least it would be an expensive idea. And the OPA1632 would then be used single ended. You need a nice fet-input opamp like the OPA134 for this job (there is a dual version the OPA2134). http://www.ti.com/lit/ds/symlink/opa134.pdf
If not silly
Yeah I don't see it being specified there but from the input bias the OPA1632 seems to be BJT.
Then what about inverting input stage with active filter and non inverting integrator so that we only use one capacitor on the servo loop instead of two?
This is leading no where, what is the price of a 1u MKP for the servo? Just do it as schema'ed or don't and make your own servo, I have a lot of time in these puppies. And they are good puppies
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.