I want to run my "direct coupled" Plinius SA-102 directly from a Mytek Brooklyn as a preamp but their help desk guy says: Brooklyn's have DC at output - our norm is at max 60 mV. 

I have no idea if that's too much at 60mV for the Plinius, or if "no DC" means "No DC at all", in which case I will be looking for a preamp option to go between the Brooklyn and the SA-102.

Any advice?

Hi paulsaints - now that you have a max. figure of 60mV from Mytek, perhaps the next step is to ask Plinius how much 'DC Offset' the SA102's input will tolerate?

DarkLantern blog - darklanternforowen.wordpress.com

I have tried that a couple ways but haven't had a reply so was casting the information net a bit wider.

If the Plinius is a true direct coupled design, ie, it amplifies dc as well as the audio signal, then 60mV dc from the preamp is most likely going to make the Plinius's protection trip. If it does amplify dc, 60mV is going to turn into perhaps 2 volts, unacceptable to be sent to a speaker. It needs to be under about 100mV from the poweramp.
If the Plinius, on the other hand, has some sort of servo loop to keep dc away from the speaker, you might be able to get away with it.
60mV dc seems awfully high for a preamp output anyway. Has it been measured to be so? It can be measured with a multimeter. Otherwise you could put a capacitor in series with the input of the Plinius to isolate it from dc from the preamp.

AMR-iFi R&D

Yes, nice to avoid a big ugly coupling cap if you can

DarkLantern blog - darklanternforowen.wordpress.com

I haven't measured the Brooklyn but the 60mv came from the Mytek people as an "average". I hear that that is quite high.

Owen - having just found a pre with "a couple of nice big coupling caps", why would I now avoid those?

and Colin - what would putting a capacitor in series with the input look like physically?

Thanks.

paulsaints - I meant that a coupling cap would usually be placed at the output (and input) of (most?) preamps in order to block any DC entering an amp cct.
And the cap would need to be a big one in this case, because in combo with the 47k input impedance of the SA-102, it becomes a high-pass filter.

(IMHO all caps are audible & the fewer the better where possible.)

DarkLantern blog - darklanternforowen.wordpress.com

The advantage of having direct coupling in a circuit is that capacitors and their attendant distortions can be eliminated. As Owen says, the fewer caps in the signal path the better. But if there must be a capacitor there, which is mostly unavoidable in valve circuits and a lot of solid state preamps, then it’s best to have a high quality one. The best ones can be expensive but a cheap one would work just as well, but you might hear the loss in quality.
With the Plinius’s input impedance of 47k a 1uf cap would roll off the bass -3dB at 3.4hz. You could put the cap inside the Mytek in line with the output rca socket (if it will fit) or put it inside the Plinius at the input rca socket. If you don’t want to open up either unit you could make a small box with the cap in it to connect in between the Mytek and the Plinius. Or have a tech do that.
Alternatively you could use a 1:1 line transformer to do the same thing, but then you will have the sound of a transformer in the signal path. Some people use transformer volume controls (TVC) which would also isolate dc from the preamp to the poweramp and be a high quality, alternative way to control volume.
But before you do any of that I’d attempt to connect the Mytek to the Plinius directly and see if there is a problem with dc. If you have a multimeter handy put it on the 2 volt dc scale and measure the voltage across the output terminals of the Plinius before the speakers are plugged in. It should be no more than 0.1 volt (100mV).

AMR-iFi R&D

sonically a good DC amp (eg. the spectrals) is like nothing, ultra clean with incredible bandwidth eg 0hz to 2 mhz.

i've heard plenty of plinius amps and am 99.9% sure they are not truly direct-coupled.

regardless of amp, 60mv of dc offset is a fair bit.

as per colin, try it first to see if there is a real problem.

My Sansui BA-F1 poweramp, like a Spectral, is a true direct coupled amp and also amplifies dc to over 1.5MHz! It was one of the first amps to use fast output transistors called LAPT (Linear Amplified Power Transistor) made by Sanken. The construction of one of these sees many smaller, fast transistors in parallel each with its own small emitter resistor inside the one transistor case. Spectral uses naturally lightning fast Mosfets. The Plinius is likely to have LAPT output transistors as well. Most modern amps use them. Having a fast amp means the negative feedback can operate with almost no phase shift in the audio bandwidth so distortion is kept very low. The Sansui has both an ac and a dc coupled input to select from. If the Mytek's 60mV were connected to it on the dc input, the protection would be sure to trip.
If the Plinius doesn’t amplify dc it should be ok to connect to the Mytek.

AMR-iFi R&D

spectral uses v-fets (like some classic sonys !)

how does a v-fet differ from a mosfet ?

Vfets? I saw some Spectrals in the CES in Las Vegas (a while ago now!) and they had Hitachi 2sk1058 and 2sj162 lateral mosfets in them. Do modern ones have SITs (Static Induction Transistor) in them?
A Vfet is Sony's name for the SIT, named because of its vertical internal construction as opposed to lateral, like the Hitachis. (Some mosfets are also constructed in vertical configuration to get the transition time of the transistor as short as possible). A Vfet is a depletion mode transistor that operates like a power jfet, with output characteristics that roughly resemble those of a valve triode. Mosfets are enhancement mode devices whose output characteristics roughly resemble those of a valve pentode. They are both lightning fast devices.

AMR-iFi R&D

older model but they say v-fet

www.spectralaudio.com/dma360.htm

nelson pass was so fond of the sony v-fet that he scoured the planet for enough to build a single pair of 40th anniversary v-fet amplifiers.

www.stereophile.com/content/pass-labs-pays-tribute-sony

By 'direct-coupled', what does that mean?
Do we mean DC-coupled, not AC-coupled, ie. without capacitors in the signal cct from input through to output?

DarkLantern blog - darklanternforowen.wordpress.com

Very interesting, Spectral use custom Vfets. They look like modern SITs and would be quite rare and exclusive.
I heard about Nelson Pass's homage to the Sony Vfet, a cool amp! He tried a version with modern custom made SemiSouth SITs but unfortunately SemiSouth went bust and the modern SITs are no more! I wonder how Spectral are doing with their Vfet/SIT supply. The production run of the DMA-360 was limited, probably due to the problematic supply of Vfets.
When I listened to Rocl's Sony TA 4650 I was hooked. Unfortunately his amp suffered a serious malfunction and now it is in Melbs awaiting repair. I have some Sony Vfets here in the UK with which to repair it. And enough to make a Vfet amp.
I suspect the demand for Vfets/SITs from audiophiles looking for the Rolls Royce device is only going to escalate, enough that a major manufacturer would make them again. I could see Toshiba or Sanken making them if there was enough demand. But high end audio is a small industry compared to computers. Although if Sanken still make LAPTs for normal amps, I wonder if they could see it in their hearts to make Vfets again.
Direct coupled circuits, of course.

AMR-iFi R&D

Owen, yes I’m taking direct coupled to mean able to amplify dc, or dc coupled if you like. That precludes the use of capacitors in the signal path.
Some direct coupled amps have a servo loop, a sub-circuit with an opamp and capacitor that keeps the dc output of a direct coupled amp under control; so that at dc, the circuit has a gain of 1 or less. The servo loop has a capacitor in it that isn’t in the direct amplifier signal path. That way the current through the servo loop capacitor can be kept low enough that it doesn’t generate much distortion, and the main amp circuit can operate with low levels of distortion.

AMR-iFi R&D

You can see the servo loop in this extract from the Luxman L580. It is direct coupled, but the gain at dc is zero. The servo loop capacitor is C708a.
I'm assuming the Plinius has a system similar to this.

AMR-iFi R&D

Thanks for the discussion. I was in CHCH today and dropped in to Plinius and may report a bit more later, but the circuit they were able to find for the SA-102 showed that Series 3 had a servo loop (I presume these are at the amp output end). What they weren't sure on was whether all SA-102's manufactured were "Series 3" or whether the SA-102 went through three manufactured series. When I get home I'll open it up to see if the series number is there but my amp is a low serial #1294. They also suggested connecting the Mytek and Amp up without speakers and checking the DC out. They agreed 60mv was vey high and might put 2v (I think) through the speakers if a servo was not in play.

Knowing that the input is 60mV DC, and that one would be relying on the servo to permanently protect the speakers from a 2v dc input, would you actually take the risk of depending on that servo never failing?

It’s not ideal to have 60mV of dc at the output of a preamp (Why so much? Ask Mytek....maybe they could consult with me to help them to get it down?) You could rely on the servo loop to keep dc under control but only if it has enough capacity to correct it. Most do. If I saw the circuit of the Plinius I might be able to make some better recommendations.
If you have the Mytek there, see if you can measure the dc on the output terminals to see that 60mV is the case.

AMR-iFi R&D

The Sa102 is DC coupled at its input. The amp uses a servo to ensure minimal dc at its output and any DC at its input will cause the servo to try to make corrections. This will throw things off.

The Plinius power amps are or were all like this. They assume that any good source will be free of DC, as they should be.

A good capacitor in series with the signal to the amp will fix this, spend as much as you want to. If the cap is in the DAC it should have a 1 megohm resistor across the output rca connector to ensure the caps output doesn't float with charge.

I know this because I designed those amps. There are none of the original staff left at Plinius, I am guessing that there is little knowledge there of the earlier designs.

____________
Pureaudio
www.pureaudio.co.nz/
www.facebook.com/pureaudio.co.nz/

Thanks very much, Gary. That's moved us from guessing to hard info.

Having learned quite a bit over the last few weeks I have cut my search for a tube pre-amp down to one with coupling capacitors on the output, or a SS M8, when one comes up.

Should have picked up that Pure preamp that was on Trademe a while ago ...

Don't buy a Plinius preamp. The power amps were great, but the preamps never reached the same standard. That includes anything I designed. I suspect the newer models are just reusing the circuitry I designed some 15 years ago..

Oct 6, 2018 15:25:06 GMT 12 paulsaints said:

Thanks very much, Gary. That's moved us from guessing to hard info.

Having learned quite a bit over the last few weeks I have cut my search for a tube pre-amp down to one with coupling capacitors on the output, or a SS M8, when one comes up.

Should have picked up that Pure preamp that was on Trademe a while ago ...

____________
Pureaudio
www.pureaudio.co.nz/
www.facebook.com/pureaudio.co.nz/

Wow! What were the compromises in the Plinius preamps?
Garym, when you say connecting a preamp etc with a small amount of dc at its output to a Plinius poweramp will throw things off, do you mean than the input differential pair might become slightly unbalanced at dc and the distortion might rise?

AMR-iFi R&D

I bought a Plinius M7 to try out.Compared it to my Clone Note LDR "passive". Very close in sound,with a cd feeding into both out to a Musical Fidelity XP200 which has two inputs,so easy to swap between the 2 while playing.
No loss on bass or anything with the Clone Note.

Then I removed the CN and connected the M7 up to my system to have a longer listen.After a week I had to swap back to the CN. What a relief!The M7 has a steely edge to it that makes listening painful on violins and male voices!Like going from the Dynavector P75/3 to the LV-1!

Wonder if there's a output cap I can change to help,otherwise I might have to move it on.

If you are handy, take a look inside at the PCB behind the output sockets.
You may find a big coupling/DC blocking cap connected in series with each RCA centre pins.
It might be something like 1uF to 3uF, either a big film cap or a small non-polar electrolytic.
You could try swapping this - but unless the bott panel is removable, the PCB may need lifting...

DarkLantern blog - darklanternforowen.wordpress.com

Yeah, must do that Owen.Is that an M5? Can't make it out.Mine has Wima caps throughout.When I replaced the Wima input caps on my Perreaux 2150B it made a massive difference.Came alive!!

Have had the bottom off as well as the top. Take photos on the phone now so I can ponder on the parts!

Have an SM3 as well. One day I'll replace the signal caps.But the M7 is not as deep as the SM3, which helps with the Antipodes cables,which are heatshrunk quite a way out from the plugs,and won't bend!