The Mote part looks very nice and it’s cheap. I have a preference for single ended JFET MM stages (lower noise). If you parallel the two JFETs in the dual package you easily have 0.5uV noise device. If it’s for low noise HF applications like AM tuners it’s probably low noise - like the old Philips/NXP BF862 (now discontinued) spec’d at 0.8uV typ.
The differential stage in theory has a 3 dB higher noise level. But if other sources of noise are taken into account, the degradation will not be so significant. For applications such as phono stage, the signal-to-noise ratio will be sufficient. On the other hand, the differential stage has good linearity, high PSRR, and it easily allows you to set the operating current using a current source. For my transimpedance phono stage, I haven't decided yet which input stage to use. It is possible to use a single JFET at the input together with a BJT in the form of a cascode circuit. Even better is a folded cascode to allow different transistor currents to be set for optimum gain. But it is not entirely clear how to adjust the operating point. I would not like to add something to the source circuit, it is better to control the supply voltage of the load resistor in the drain. But the initial drain current is too high. While a diode in the source circuit might help...
Last edited:
If you can source all the transistors, I bet you won't regret making Scott Wurcer's preamp as it seems he also solved the hard clipping problem on dusty records with his topology.Unfortunately it's a pretty expensive solution for the normal diy-er:
https://www.diyaudio.com/community/...erential-or-single-ended-phono-preamp.306811/
EUVL says 2sj74 could be replaced with bipolars in the cascode, but i have the feeling that the soft clipping will be lost with his transistor.
An older suggestion for a bipolar transistor prone to quasisaturation , and maybe a better clipping behavior would be mpsa18 :
https://www.diyaudio.com/community/...saturation-measured-data.293308/#post-4755597
Last edited:
Thank you, but I'm not interested in building a ready-made preamp. The whole point is to develop something of your own. I don’t like the circuitry at the link above, I like more modern solutions using opamps. I can take some fragments, but not the whole schematic.
It is RF component, - noise is specified @ 100 MHz and no DC offset data... I don't want to buy a pig in a poke
Indeed, on an MM preamp there is a point at which lowering the input noise adds no practical benefit because of the transducer noise effect and the source material - better to have a balanced approach that takes care of noise in a practical way and also addresses the dynamic requirements of vinyl. But, there's always the engineering challenge of course 🙂
Pitty they didn't test 10kHz filter. Perhaps they would find out 10kHz limited audio is the best 😉
As you like. For example, BF862 is also an RF device. But this did not prevent its use in low-frequency circuits. The noise graph of the CPH3910 is close to that of the LFU2140. The only drawback is that the match is not normalized for the CPH6904 pair.
The BF852 is spec’d at 0.8uV typ. It has very low noise at LF and a low 1/f corner. They get used in mic preamps quite often. Scott Wurcer was very impressed with it.
The BF862 datasheet only says a noise density of 0.8 nV/Hz^0.5 at 100 kHz. There is no data on noise in the audio range, and there is also no cutoff frequency of 1/f noise.
Attachments
They’ve been measured - one of the guys active here on the forum published measured results for a number of JFET’s in IIRC ‘Electronic Design’. The BF862 came out tops by a good margin. It’s an ‘industry secret’ BTW - the go to low noise JFET for audio.
I know that BF862 has been used successfully in audio, but their production has been discontinued. Horowitz and Hill name the CPH3910 as a replacement.
I didn’t read this forum very much (it’s hard for me to do this, I understand technical English, but a lot of words I don’t understand are used on the forum), but sometimes I saw his messages, it’s clear that he is an experienced person. But the amplifier circuit, consisting entirely of JFET, is not a product of our time. But I promise to read more about that amplifier.
Is this chip op amp modern enough for you?
https://www2.mouser.com/ProductDetail/Analog-Devices/AD797ARZ?qs=/tpEQrCGXCyVh/EXLVBgtw==
Scott designed it.I don't write random recommendations.You wanted a folded cascode design, i recommended you the best low noise, low thd folded cascode designer...
Thanks, that's cool! But I don't need to be that good. It's not about achieving some record parameters, but about a fresh and interesting idea. So far, the only phono stage from this site that has touched the soul is Aurak.
Right, a fresh ideea applied to the old vinyl...how fresh do you want it? One week, one hour?And how good if too good is not good?
https://www.labjc.com/?p=1179
https://www.labjc.com/?p=1179
As the schematic is created, questions will arise. Then I'll be ready to ask them. It's probably better to do it not in this thread, another amplifier is being discussed here.
Please take your time and explore the linked threads.
https://www.diyaudio.com/community/...d-posts-in-analog-sources.104832/post-5714735
They are all a good technical read. (Joachim’s is a treasure box).
My Aurak trimmed for the M97xE is
https://www.diyaudio.com/community/...orch-preamplifier-part-ii.146693/post-5172604
George