• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Schmitt dephaser

Status
Not open for further replies.
In the description of some of Jadis amplifiers I found the following statement:

"A modified Schmitt dephaser at the input creates a crossed and self-balancing feeback fro the ECC 82 tube"

www.jadiselectronics.com/pages_eng/produits/menus/menu_amplis.htm

Could someone explain me what the Schmitt Dephaser is ?

I only found a schematic of one of those amps

http://www.pi.infn.it/~federico/Immagini/jadis.zip

, and the Schmidtt dephaser looks like a very poorly ballanced - through a 82 ohm ressistor in the "long tail" differrential amplifier, where the second tube (with the grounded grid) gets only a very smal signal from the 82 ohm ressistor and provides a small global POSITIVE feedback to the amp.
 
Could someone explain me what the Schmitt Dephaser is ?

Real bad English. There is such a thing as a "Schmidt phase inverter", a.k.a. Long Tail Pair a.k.a. Differential amp. It's simply a common differential amp that's used as an unbalanced-to-balanced stage as an active single-to-PP xfmr. It's actually one of the better active phase splitters (the cathodyne being the other one).

the Schmidtt dephaser looks like a very poorly ballanced - through a 82 ohm ressistor in the "long tail" differrential amplifier, where the second tube (with the grounded grid) gets only a very smal signal from the 82 ohm ressistor and provides a small global POSITIVE feedback to the amp.

Good eye, there. Furthermore, it's a 12AU7A which has an amplification factor of no more than 20. Now, you might get away with using a very small resistor in the tail of a BJT LTP simply because BJTs have way more gain than a 12AU7A. Even with BJTs, 82 ohms in the tail would be problematic. He makes that bad situation even worse by including cathode degeneration resistors that are bigger than the tail resistor. This is a spectacularly bad design, and I would avoid it like the plague. Oh, and let's consider what a poor audio tube the 12AU7A is: the linearity plainly SUX, this being more of an RF and digital tube.

Oh, and the schematic given in that zip file is incomplete.

So far as this design goes: FUGGEDABOUDIT! :whazzat:
 
I suspect the 82 ohms is a misprint for 82K ohms.

It isn't. The schematic gives some values for current and voltage. He's got that 12AU7A LTP connected to +375V(DC), with an I(P)= 3.0mA. Given the values: (100E3 * 3E-3) + (474 * 3E-3) + (82.5 * 6E-3)= 301.971V(DC). That agrees with the voltages given, leaving a V(pk)= 73.083V(DC). There is no way that that's a misprint.

What he does with the second stage makes no sense at all. Here, he has a 49.7K resistor in each cathode of the paralleled 12AX7As, bypassed with a capacitor, and a tail resistance of paralleled 47.9K's. WTF?! :bigeyes: Why doesn't this stage have one tail resistor?

Once again: bad design. :whazzat:

Avoid. :angel:
 
So the word "dephaser" must be of French origin and simply means a splitter.

Another strange thing about this amp is that it uses 2 ECC83 /12ax7 with a 330 kohm plate ressistor to drive 3 parralel kt 88 OT ! instead of cathode followers or low Ra tubes for example.

I also agree that this looks like a bery bad design but all Jadis amplifiers are generally respected and cost thousands of $
 
I also agree that this looks like a bery bad design but all Jadis amplifiers are generally respected and cost thousands of $

The audio field is just full of all sorts of con men selling some of the most unbelieveable things for big $$$$: The $500 Volume Control Knob (I'm not making this up.) There's lots more where that came from.

Then there's the psychological aspect: there are very few folks out there with the fortitude to admit that they were had. So, regardless of how bad the design of the Jadis amp is, you will find lots of folks out there praising it simply because they don't want to admit, either to themselves or others, that they wasted "thousands of $".

It's still a very bad design that looks like it was designed by someone more familiar with solid state designs. He should have either stuck to doing "sand", or done his research on hollow state before attempting this. But what the hey, if he's getting "thousands of $", then what does he care? 😀
 
I also agree that this looks like a bery bad design but all Jadis amplifiers are generally respected and cost thousands of $

Perhaps, but I have also heard an undercurrent suggesting just the opposite for Jadis’ reputation in the design department. I have heard others talk about the “flaky” designs of Jadis in the past. This is one such design, unless the schematic was copied very wrongly. I have to agree with the prior posts: This design is really pretty bad on several counts. I have to wonder what these guys were thinking – were they even technically trained? It is a testament to vacuum tubes (especially with some negative feedback to tidy up the mess) that even when used sub-optimally, they can sound at least “OK”. But, some people do like the sound of lots of second harmonic distortion, which, along with lots of third harmonics and some bandwidth limitations, this amp will deliver in spades.
 
What he does with the second stage makes no sense at all. Here, he has a 49.7K resistor in each cathode of the paralleled 12AX7As, bypassed with a capacitor, and a tail resistance of paralleled 47.9K's. WTF?! Why doesn't this stage have one tail resistor?

Precisely, and the decoupling caps on those cathode resistors also have reversed polarity. This shows lack of quality control in design (if the word "design" applies in this case).

I'm not at all familiar with Jadis but I gather from other people's comments that it's a reasonably well-known brand with a respected reputation. How can that be, with such poor designs? Already mentioned are the 12AU7 tubes (agreed, they're not the best choice - not as bad, though, as 12AT7) and the weird resistance values used in the amp itself, but the power supply isn't the greatest, either, is it? It all seems very amateurish to me.
 
How can that be, with such poor designs?

Fashion. They're shiny! They're French! They're as tempermental as a sports car! They measure badly! (take THAT you meter-readers!)


Already mentioned are the 12AU7 tubes (agreed, they're not the best choice - not as bad, though, as 12AT7)

???

In voltage amplifier service, the 12AT7 measures quite well, much better than the 12AU7. The word-of-mouth bad reputation of the 12AT7 still astonishes me. Of all the 9-pin miniatures, it's probably the best commonly available tube for a power amp input stage.
 
Sorry, I was deliberately stirring, trying to find out if there were any ECC81/12AT7 fans around! Unless one has direct experience of it (which I don't), one's opinion of this tube is likely to be influenced by the writings of others, most of which has been uncomplimentary.

Without naming names, there are those who are inclined to disparage any tube designed for RF (such as 6DJ8, ECC81/12AT7, ECC82/12AU7), on the basis that such a tube wasn't designed for and, therefore, cannot be any good for, audio purposes. Some perfectionists automatically turn their noses up at anything smaller than octal; indeed, there are some rare individuals to whom even octal constitutes a slide into the abyss of mediocrity. and as for pentodes, well . . .
 
Sorry, I was deliberately stirring, trying to find out if there were any ECC81/12AT7 fans around! Unless one has direct experience of it (which I don't), one's opinion of this tube is likely to be influenced by the writings of others, most of which has been uncomplimentary.

I don't tend to put much faith into that. I prefer to let a load line analysis do all the talkin'. If the THD estimates look good, then what else is there?

Without naming names, there are those who are inclined to disparage any tube designed for RF (such as 6DJ8, ECC81/12AT7, ECC82/12AU7), on the basis that such a tube wasn't designed for and, therefore, cannot be any good for, audio purposes.

Depends. 807, 833A, 813 -- all originally designed for RF, or the 2A3 (series pass regulator) and all make excellent audio amps. 50L6, 50C5, etc. -- designed for audio, and yet... 🙄 There are good non-audio types and some horrible audio types. AFAIC, let them disparage those RF types: more for me. 😉

Some perfectionists automatically turn their noses up at anything smaller than octal; indeed, there are some rare individuals to whom even octal constitutes a slide into the abyss of mediocrity.

Much truth to this, however. All pre-octals, and most octals, were designed correctly: as high voltage, low current devices. The problem occured in trying to force low voltage, high current operation. This made for cheaper mass market electronics by the avoidance of high voltage power xfmr's with the need for more copper and iron, as well as more stringent insulation requirements. With lower voltage, it's easy to substitute cheaper RC filters for LC filters, the lower impedances meant less iron and copper for matching xfmr's. It also led to linearity problems as seen with the "suicide box" finals such as 50C5's. You can find good miniatures (6AQ5 -- essentially a down-rated 6V6) or types like the 12AT7A that have good THD figures.

Best to do your own research, and not bother with all the hype (especially if you suspect that someone is trying to sell you something. 😉 )
 
I don't tend to put much faith into that. I prefer to let a load line analysis do all the talkin'. If the THD estimates look good, then what else is there?

You're right, of course. It's amazing, not to say amusing, how the "experts" disagree.

Depends. 807, 833A, 813 -- all originally designed for RF, or the 2A3 (series pass regulator) and all make excellent audio amps. 50L6, 50C5, etc. -- designed for audio, and yet... There are good non-audio types and some horrible audio types. AFAIC, let them disparage those RF types: more for me.

Yup, nothing wrong with 807, 6146, 6AU6 and a few other RF stars (hey, this colour thing's fun!)

. . . not bother with all the hype (especially if you suspect that someone is trying to sell you something.

I hate when that happens.
 
ECC82 (12au7) in my opinion is not the best choice for audio aplications, and I'm quite suprised so many companies still use this tube as a driver. Maybe there are more linear versions. I never was able to get low THD both at low and high levels. Ecc81 (12at7) is far more linear. My favourite tubes are however Russian 6N1p-EW (and only EW-army version) and Philips e88cc-sq both for low-signal stages and drivers.
The fact some tubes are designed for RF is not important as far as the other parameters are OK.
 
Status
Not open for further replies.