Simplistic NJFET RIAA

diyAudio Chief Moderator
Joined 2002
Paid Member
Many people seem to prefer transformers. Can good Cinemag, Lundahl, etc. really be that much better than these Shure transformers?

Anyways, JFET mc pre will stay for now. I will concentrate my efforts on power supply and maybe try the Hiraga BJT mc pre for comparison.

There are differences. Not tragic but palpable. Best I had tried is Sowter*. Symmetrical impedance SUTs are very good for balanced signal transfer from a naturally floating transducer. Like the MC cart is. Not cheap. When connecting SE, and if a diyer, some high quality active pre pre is a good bet. *https://www.diyaudio.com/forums/ana...matic-balanced-input-riaa-17.html#post5721652

P.S. Be careful to always check for zero or near zero input DC offset on both channels before connecting the TT when trying BJT pre pre circuits. Also see that they stay trimmed with ambient temperature changes. Accidents during development if DC input coupled can kill a cart. Forget the JFET input safety mindset.
 
It has two practical advantages over the Denon. Higher output and a replaceable stylus. High output is good on high sens speakers. Not to crank the phono gain or the vol pot and bring up the system's noise floor. In general the DL-110 is velvety, the MP-150 is heftier sounding and more resolving.

This guy loves it. He is fun to watch.

YouTube

Thanks for the recommendations. Just as an update, I was feeling fancy and bought the Nagaoka MP-200. It is very impressive and I have really been enjoying it. One thing that is disappointing is when some of my records (mostly the newer ones) have bad mastering jobs. It is a shame they couldn't do a better job and make the record sound better than the digital files. I remember when I set up my subwoofer for the first time and discovered what songs handled bass well and which didn't. This is kind of similar with quality of the media, getting to find out which records are high quality and which are not.
 
I have been experimenting with this step-up...which has turned out well. Mine is built on perfboard with PRP resistors, Nichicon KZ and a Mundorf Supreme output cap. So far, I prefer it to transformers, although I haven't tried big $$$ transformers.

This circuit is very sensitive to power supply quality. I have tried LM317, LM317 with adjustment cap and a variety of batteries.

LM317 with an external transformer (12VAC wall wart) is not so good. It's OK, but noise is an issue and it lacks some sparkle. Probably still better sounding than my mic transformers, but noisier.

LM317 with adjustment cap is better. Got some sparkle back. Still noisy. Not terrible, but not as good as batteries.

Set of 6 AA batteries sounds good. Less noise, more elastic sounding. Carbon batteries not as good as alkaline.

2 sets of 6 AA batteries (dual mono) sounds really excellent. Especially if the batteries are alkaline. Really liquid sounding, least amount of noise.

2 sets of 6AA Batteries (dual mono) followed by a pair of 33,000uf caps (like Hiraga would use)....this is really nice. Best of everything. Power, control, blackness and very liquid. Sounds very rich and free.

Hiraga must have known what he was doing. And this is a very nice circuit if you are willing to experiment.
 

Attachments

  • Salas step Up.png
    Salas step Up.png
    595.2 KB · Views: 364
Yes. And cap multiplier after batteries as well. I did not try 7809 because I don't have any.

Linear PSU was 12V AC wall wart, MUR160, 10K UF Nichicon KG, LM317. Single mono.

In general, I hate batteries. I've used the nfet mm phono with batteries. Salas shunt was clearly better. But in this case I will make an exception.
 
I have read tests where batteries are compared to linear psu. Noise in batteries increases greatly as current draw increases. Perhaps that’s why batteries perform well in this situation. In other circuits I have not preferred batteries.

Also, I want to be clear that this circuit, regardless of power supply is good and I prefer it to transformers. But Iam crazy and always have a soldering iron ready for experimenting.
 
Last edited:
diyAudio Chief Moderator
Joined 2002
Paid Member
Its a sensitive zero PSRR very low signal point. That asks for little current and it can use the lowest noise rail it gets. Batteries are not associated with the mains EMI too. When you want to extract all its potential, PSU perfectionism as well as shielding, matters.
 
Yes, and my environment is not best place for phono stages with a big class a amp and huge torroid transformer less than a meter away. Grounding also matters greatly in this circuit, and Iam using a rega which complicates things. I have tried a few grounding schemes but it seems to work best bringing all grounds to the rca jack the rega uses for ground. Regas have no separate ground wire.

But, as usual, all Salas circuits sound good. This is more tweaking and testing for my own personal environment and situation than anything.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
In general, I hate batteries. I've used the nfet mm phono with batteries. Salas shunt was clearly better. But in this case I will make an exception.

Yes, and my environment is not best place for phono stages with a big class a amp and huge torroid transformer less than a meter away. Grounding also matters greatly in this circuit, and Iam using a rega which complicates things. I have tried a few grounding schemes but it seems to work best bringing all grounds to the rca jack the rega uses for ground. Regas have no separate ground wire.

But, as usual, all Salas circuits sound good. This is more tweaking and testing for my own personal environment and situation than anything.

Yes, I know the practical difficulties you are facing. After we gather enough experience and we are careful, an MC pre-pre can be executed well for low self noise floor with almost absent rail noise influence and tiny harmonic spikes though. Even when conventionally mains powered.

Here is a custom one I made in 2012 for an audiophile in Germany. He had a VDH Colibri XPP MC cartridge with platinum wire coils which presented a weird output impedance due to the higher resistance metallurgy and their four layer winding inductance that he simply could not find a SUT to match well although he had tried.

He was looking for a SUT to start with because when hooked up directly to the MC input of his very expensive phono stage it was sounding like a mediocre CD he said. A batteries powered German design with sophisticated recharge automation. Possibly an ASR high rank model, can't remember well, its been years. Maybe an RFI problem due to inductance and non optimum termination I suspected. And/or underperforming BJT input type op-amps MC circuitry when on a source impedance they were not designed for.

It was a special MC cartridge of high cost and great sonic potential but with an unexpected signal engine for its category in other words. Difficult to predict with what hi-end gear to mate best. It had fairly high output at least. Nearing 0.4mV.

Make something for me of your inspiration that it may finally suit my cart and I am willing to sponsor it even as a failed experiment he said. So I did devise and built something for a very fair charge. Nobody was 100% sure of how good it will mate with the whimsical cartridge, but it all went very well in the end as he replied with happy results when he got it.

Later on he asked for a BiB 1.1 external PSU also to replace the linear wall pack. He reported some additional subjective gains for including that one too.

The FFT result you see is for 0.5mV input 5mV output when using a linear wall wart, 9V set LM317, two stage capacitance multiplier. Just careful proto-board build, not even using a groundplane of any sort (!)
 

Attachments

  • PrePreFront.jpg
    PrePreFront.jpg
    264.5 KB · Views: 375
  • PrePreBA.jpg
    PrePreBA.jpg
    375.6 KB · Views: 383
  • PrePre1.jpg
    PrePre1.jpg
    513.5 KB · Views: 344
  • PrePreFFT.png
    PrePreFFT.png
    24.6 KB · Views: 427
  • PrePreCRKT.png
    PrePreCRKT.png
    12.9 KB · Views: 433
Good Day All,
Well not really, I seem to have had a disaster and was looking for some advice. I was trying to switch pre-amps in my system from the WHAMMY headphone amp I've modified to a Schitt Vali just to give my ears a fresh perspective when there was a brief light show that went from the Vali, from what I can tell, backwards to the Salas Phono Stage. The audible after effects are a garbled signal switching channels dropping out and reappearing all the while varying in amplitude. The Power supply of the amp is still fine, but reading DC voltage across components in the amplifier portion gives me non symmetrical and undulating readings. What do you think happened, first of all, and what are the most likely candidates for having been fried. Where to start?
I appreciate any help or guidance in this trying pisser of a time.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Can't guess a reason. Was it a live switchover? Examine the Vali for zero Volt DC across its inputs/outputs. Is the signal situation still the same when putting back the WHAMMY? Closer to output are Q5 Q6, so more likely candidates. List DC voltage measurements between drain and ground of all JFETs per channel so we can start putting together a picture.
 
Thank You
Not a live switchout, it became an issue at power up on the Vali and I think it originated in the tube, signal is the same with the WHAMMY, power amp was connected and I thought I saw fire in the chassis, but it appears undamaged, i.e. still makes noise. Vali DC at ins and outs at a moderate average of 17mV and present in all.
Will advise on the voltages requested ASAP.
You are now and have been most helpful, sir.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
There is some conflict here i.e. the rail branches have correct voltages at their local filter caps like normal bias current is pulled by the semis but their bias voltages are dead as you reported in mV figures their drain to PSU zero point checks.

Q3 transistors could have been hit so some weirdness occurs, see if they got normal Vbe (about 0.6V). Also what are the voltages across R4's?

For a quick check on JFETs, before desoldering any to verify which work or not by doing IDSS measurements, while power is off you can measure their RDS. Its the Ohmic value between their drain and source pins. 2SK170BL may roughly show 30-50Ω, it changes with ambient temperature and length of measurement, 2SK369BL about half that. Even less when doubled in parallel, but including a source resistor value when there. Both types when healthy certainly must not show a near short neither hundreds of Ohm.