UNU-pnono riaa mm preamp

[dreamth]

A year and a half ago I started to try something for my personal use.After reading allmost everyting I could get on this subject,and also after about one year of trials this is what I've done.

http://i61.tinypic.com/34t72hw.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
No other noise than OMRON counter source , which is at the lowest audible limit.Of course, that counter can be removed, but it's noise is almost inaudible .
I use 200 ohms old Universum headphones over the line resitor ladder for line output and the sound is gorgeous, although the output greatly supports 32 ohms headphones.I have inside even a second solid state headphones preamp for low impedance headphones, inspired from A820 Studer Revox line and headphones preamp and the tube output is clearly better with 200 ohms headphones.I suspect that If I could do replicate the original solid tate A820 Studer revox line-headphones preamp with that OTP that I missed, the tubes could loose the battle, or could be even match..

this could be a follower to UNU preamp ("one" in english):
DOI(it means "two"):
http://i57.tinypic.com/9bc5j6.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

Of course Alan Kimmel, Stuart Yaniger , Greg van Sluys, Patrick Turner, Tektronics guide lines about scopes vertical preamps, even kenwood l-02a solid state architecture and many other men and net stuffs guided me step by step.
For me, I'm a solid state follower for ever, but I tried to understand what I can be done to make tubes sing .
Of course, with proper pcb design and other superior materials the optic aspect of the beast can be improved.
And these are the inside and outside pics of the monster:
http://i58.tinypic.com/fxcx7a.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i57.tinypic.com/2cnjq8l.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i58.tinypic.com/sz7jie.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i60.tinypic.com/33kudki.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i62.tinypic.com/33de009.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i60.tinypic.com/30k8yee.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i59.tinypic.com/2ldax3b.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i61.tinypic.com/142v2wi.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i62.tinypic.com/28i9rw0.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i57.tinypic.com/14chg5v.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
http://i62.tinypic.com/11ry6hk.jpg
For mc cartridges I have my own transistor cart, in common base cascoded architecture with 16 pcs 2sa1038 selected from factory(pink spoted) for very low noise.In respect for bjt qualities, I am convinced that no tube or high transconductance fet what ever could it be , can be better tha proper paralleled bjt input.For any ss cart I have another inside stabilised source with +- 18, 15, 12, 8 an 6 volts which can be seen in some inside pictures.The radiator wings above the "Thing"(between tubes) have a mje350 as the upper possible cascode ccs that could replace the one half e288cc ccs'c upper cascode .For now there's no need .
The input silver mica capacitor is composed from 100pf and 330pf removable capacitor.For better cartridges that the one I used for tests(one old Onkyo, shure m75d and philips500), I'll make in the future an external support for various input dumping capacitors.

 

[dirkwright]

So, literally, back-of-the-envelope designing? Glad it works well for you.

 

[dreamth]

Back-of-the-envelope work too...
I wanted to save the e810f and without having matched fets for thermal compensation, I just did it this week-end with BJT which I have a lot.It works well for my ears, but again: no SPICE, no advanced calculations, no true measurements....With one channel transistor and one e810f I listend Jim Horn and some others on a Siemens RW777 with shure m75d and onkio cartridges and felt perfect for two hours with less hot glass.

An externally hosted image should be here but it was not working when we last tested it.

http://i59.tinypic.com/10rmvr4.jpg


I had no available pcb , other than two LEM input preamps from a mixer so I used them:


http://i60.tinypic.com/nzo93b.jpg
http://i62.tinypic.com/2zf08hs.jpg


I never understood why for high transconductace tubes like e 810f or d3a the plate voltage would have to be higher than g2, in fact I used even 80 volts less anode voltage for pentode configuration without any worse effect so.... I just tried transistors with 7 and 15 vce and it works coolest than ever without any perceptible degradation.In fact I think there is no need for tube followers at all in preamplifiers if we have transistors.

 

[dreamth]

last schematic of the input configuration:

An externally hosted image should be here but it was not working when we last tested it.

http://i62.tinypic.com/33xwsaa.jpg
I don't remember exactly the riaa componenents.

first measurement when I got the scope:

An externally hosted image should be here but it was not working when we last tested it.

20khz

last output measurements with anti-riaa(Patrick Turner) and riaa filter involved-there is some nonlinear transfer characteristic for the low frequencies but i don't know exactly if the anti-riaa ,riaa or both filters are guilty.The measured output signal is the upper one on the scope and last on paper.The lower signal on scope is at the input of anti-riaa filter.
http://i61.tinypic.com/e7ylpz.jpg
http://i60.tinypic.com/s5a2kp.jpg
http://i61.tinypic.com/27x08rn.jpg
http://i61.tinypic.com/2m43880.jpg
http://i57.tinypic.com/zivx94.jpg
http://i57.tinypic.com/1zlu1k7.jpg
http://i59.tinypic.com/167o9yc.jpg
last measurements right after the first stage and the riaa filter:
http://i61.tinypic.com/2dufj12.jpg
http://i59.tinypic.com/1687uz7.jpg
http://i58.tinypic.com/k3kns3.jpg
http://i57.tinypic.com/es4utd.jpg
http://i57.tinypic.com/8yyohc.jpg
http://i58.tinypic.com/j8zb6o.jpg

last appearance :
http://i60.tinypic.com/aa9ehl.jpg

 

[dreamth]

I tried a new riaa passive network with lower impedance , than I made a new circuit that could change between the high impedance and low impedance riaa network.This are the schematics of the two filters:

An externally hosted image should be here but it was not working when we last tested it.

These are the new lower z filters:

http://i60.tinypic.com/317f9n8.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i62.tinypic.com/oj1hk1.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

These are the scope pics with the measurements with lower z filter on one channel and the higher z filter on the other.I suppose it doesn't matters who is who and which is which, but on the paper in each picture "SUS" means upper side on scope(lower z network) and "JOS" means down side of scope where higher z is measured:

http://i58.tinypic.com/2iw8v0h.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i57.tinypic.com/15eiiy0.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i57.tinypic.com/10z79qu.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i59.tinypic.com/2a0i9mo.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i57.tinypic.com/wmjapt.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i58.tinypic.com/123y1xi.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i59.tinypic.com/1z3p5y1.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

Again the low frequency attenuation is here because my generator can't make a fair enough deal with the anti-riaa network but I can assure you that there's enough bass , ever more than I am used to.

And this is the new look for now:
oak sweet oak

http://i57.tinypic.com/11hs5xt.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i57.tinypic.com/2gxfx37.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting

http://i59.tinypic.com/zkke8l.jpg
Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting
Of course, the rope is the same vintage cotton rope to keep capacitors in place ...
But...the sound is slightly different in each case .With lower z fillters is less vibrant, more flat like response and i feel I listen to a good transistor preamplifier period.I made this try to know if I could rely on trying lower z rlc filter and I think I can.

 

[dreamth]

Although a bit weird to raise from the dead a topic about a phono preamp that was finished, measured and listened, after i finished my fet based second version here: 1V2 phono riaa preamplifier.
I felt that i needed to address the noise problem as my first attempt in 2014 had no issue with that , being helped by ultra low noise bipolars. To my shame in 2014 i didn't know how to use a simulator and i lost a lot of time for the first design when now it took me just two days to do the work i did in months of trial and error before...
So i wanted to give that monster a second chance by simulating it for better specs if possible.I used some other transistor pairs in the sim , but bd139, mje340,bf459 instead of ksc3503 and 2sa1038, 2sca992 instead of bc550c might be some other good choices too.My original design used mje340 and 2sa1038 ...I'll probably go with bf459 and 2sa1038 or ksa992 now...
You have here some sims with two slightly different mu-follower ideas:
In the floating mu version i have 300mv more output at 0.5mv 1khz, but most probably i'll be having some issues with different operating points as high transconductance valves go wild sometimes, if they are not given some constraints and i've seen it a lot with d3a, but everything is capacitor coupled so no real worries.


The distortion pattern is raised by 15...20 db by the second stage actually .I consider using an op-amp for the second stage to make this a cristal clear sound, but a valve has it's benefits when clicks and pops come in .


Most probably i'll be giving a new trial to a proper pcb for this version .I,m not good with pcb design ...so it might look like mess as usually does, but as long as it works well, i don't care of the commercial finish.It's for my own use and i am really glad that nobody believe me when i say that this design is one of the best possible in the whole phono world.
Don't copy it, just go ahead with the usual "no sand" crap !

 

[dreamth]

I totally dumped allan's kimmel mu-follower idea as loading more the bjt has no beneffit.I just load it with a high value 100kohm resistor that simply helps at start-up.I simply used the 1v2 previous sim as a base for the older design.As you can see, i didn't even change the design name...