|
Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members. |
|
Thread Tools |
21st Sep 2007, 4:32 pm | #1 |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Making a Triode Loftin White
Hi,
after making the triode coupled RIAA preamp https://www.vintage-radio.net/forum/...ad.php?t=18326 I decided to modify my RL12T15 triode amp for Loftin White. http://rl12t15.blogspot.com/ I thaught this topologie is Loftin White: http://bp0.blogger.com/_oCEpds9YoPw/...ftin+White.png The thing I don't like in this circuit is the the cathode decoupling cap for the output triode. I searched for a solution that makes it possible to eleminate the coupling cap and the cathode decoupling cap and came up to the following solution: http://bp0.blogger.com/_oCEpds9YoPw/...+schematic.png This circuit solves all problems and compensates hum and signal voltage dropping at C2 as well. Now I was very curiously whether someone did something similar. I searched in the www and found nothing similar except this one: http://www.mindspring.com/~mjh850830/loftin-white/ You can download the originals from Loftin and White there: http://www.mindspring.com/~mjh850830...e/article.html I am very impressed about the work of Loftin and White. Quote from Radio News Feb. 1930: "Forget Former Conceptions on Audio Amplifier Theory" There are two people coming to show the world how to build up a perfect valve amp and it is totally ignored and misunderstood. Note this Quote from Radio News Feb.1930: "A condenser, C2, of about 1 microfarad is needed to form a local signal circuit in the output circuit including association of any suitable loud speaker as indicated." http://www.nutshellhifi.com/library/Tube_Fest_Talk.html Where is Loftin White here? Elliano Ultrapath- an incoplete Loftin White? And Mullard's 3-3? http://vintageradio.me.uk/amplifier/mullard33_2.jpg What do you think? Kind regards, Darius |
23rd Sep 2007, 6:20 pm | #2 |
Hexode
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 406
|
Re: Making a Triode Loftin White
Hmm, it seems that you have eliminated the cathode cap by introducing positive feedback from the power valve cathode to the PC86, to make up for the lost gain. An interesting solution, though I can't say I would use positive feedback in any audio amplifier- it seems to risky.
I would simply have used LEDs to bias the PC86- no need for any capacitors at all! |
23rd Sep 2007, 6:44 pm | #3 |
Retired Dormant Member
Join Date: May 2005
Location: Eastbourne
Posts: 39
|
Re: Making a Triode Loftin White
Ah, the old Loftin-White..............
I breadboarded one a few years ago, and; at the time; found this article by Broskie. http://www.tubecad.com/may2000/page17.html I didn't have any trouble with mine, no matter how much I tried to destroy it! |
24th Sep 2007, 8:53 am | #4 | |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
Quote:
thanks for your reply. This is not any positive or negative feedback. There is no signal voltage "lost" at the cathode of the RL12T15. Please note this aspect: Quote from Radio News Feb.1930: "A condenser, C2, of about 1 microfarad is needed to form a local signal circuit in the output circuit including association of any suitable loud speaker as indicated." http://www.nutshellhifi.com/library/Tube_Fest_Talk.html Have a look at the Western Electric repeater amplifier. There is no need for a cathode decoupling of the output valve. Same situation in a Loftin White arrangement. Imagine C2's capacitance is infinite and it is charged, or say it is a perfect cap like a voltage source. In this case there is no signal voltage drop at the cathode resistor. Loftin White compensates the influence of the last smoothing cap to the signal with Chb too. Is this well explained? Chb makes the triode output floating on the cathode level of the output valve. This makes it possible for me to replace the Loftin White input tetrode by a triode. The supply voltage for the PC86 is much higher (ca.550VDC) than the anode voltage (ca.100VDC) and (!) 220K Ohms divided by 560 Ohms is much higher than micro. Thus a cathode decoupling is not neccessary here. An LED is a non linear component. I don't want to use these there. Kind regards, Darius |
|
24th Sep 2007, 9:30 am | #5 | |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
Quote:
http://www.tubecad.com/may2000/page17.html there are lots of so called "save Loftin White" in the www nothing to do with Loftin White, but a solution for a problem that does not exist in a well designed Loftin White. Loftin White is safe. Imagine what happens to an output valve in a set using a leaking coupling cap. Caused by the low resistance of the cathode resistor the current increases very much. I think you know this problem from the vintage sets. Loftin White protects the valve. A Loftin White is save and does not need a choke. Please have a look at the links in #1 Kind regards, Darius |
|
24th Sep 2007, 10:53 am | #6 |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
#5
Please note that the PC86 has two anode pins. This offers a nice feature, it makes a current shut down for the output valve possible. Have a look at http://RL12T15.blogspot.com The underlined words there are links to websites and pictures. This pic shows the connection of the PC86: http://bp3.blogger.com/_oCEpds9YoPw/..._von+unten.JPG And a surge arrester protects the output transformer from arcing. http://bp1.blogger.com/_oCEpds9YoPw/...ftin_White.JPG Kind regards, Darius |
26th Sep 2007, 8:00 pm | #7 | |
Hexode
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 406
|
Re: Making a Triode Loftin White
Quote:
You also have Chb, which is yet ANOTHER decoupling cap (decoupling via C2), so you appear to have made your problem of capacitors worse, not better. And LED may be a non-linear device, but compared to an electrolytic capacitor is is VERY linear. I would certainly prefer LEDs before electrolytics. |
|
27th Sep 2007, 8:21 am | #8 | |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
Quote:
C2 does not serve exactly the same purpose as the original cathode decoupling cap. C2 serves the same purpose as the B+ Capacitor in the "Conventional SE" have a look at http://www.nutshellhifi.com/library/Tube_Fest_Talk.html Please try to understand what Lynn Olson explaines there! C2 in Loftin White serves exactly the same purpose as the cap in the Western Electric (output). There is no need for a cathode decoupling cap. Importand: In a Loftin White it is forbidden to decouple the cathode of the output triode with a capacitor in parallel to the cathode resistor!!! Chb is not a decoupling cap in my application. Please not that there is 30K Ohms (!) in series with Chb. This is much more impedance than the cathode resistor(s) have. Thus Chb is not able to decouple there. In my application Chb makes the driver triodes anode floating. This makes it possible for me to use a triode instad of a tetrode (or pentode) in my Loftin White topologie. Have a look at the Western Electric (repeater amplifier) again. (scroll to the middle) http://www.nutshellhifi.com/library/Tube_Fest_Talk.html There is a cap (0.22uF) it makes the interstage transfromers secondary floating. Do you understand now what is going on there? At last the LED. Its non linearity causes harmonics, "non linear destortions". The result is an "LED sound" if you like this it is ok. An electrolytic changes its impedance with the frequency this gives "linear destortions". Thus the resistor is the best choise here and it causes no losses in my application as explained in #4. Kind regards, Darius |
|
28th Sep 2007, 4:11 pm | #9 | |
Hexode
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 406
|
Re: Making a Triode Loftin White
Quote:
In the Western Electric, the output transformer is connected in PARALLEL with the valve, and is coupled to the cathode via that capacitor. In your chematic, http://bp0.blogger.com/_oCEpds9YoPw/...+schematic.png the output transformer is connected in series with the anode in the normal way, and C2 decouples the cathode directly to the steady HT supply voltage, which, as far as AC is concerned, is exactly the same as decoupling to ground. |
|
28th Sep 2007, 4:27 pm | #10 | |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
Quote:
do you agree, that it does not make a difference on which end of the output transformers primary the current source is connected in the Western Electric? (Except that parallel feed does not make a DC current in the output transformer, of course.) Kind regards, Darius |
|
3rd Nov 2007, 4:40 pm | #11 |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White 300B
Hi,
proudly present the ECC83 300B schematic http://rl12t15.blogspot.com/2007_11_01_archive.html Thanks for looking, Darius |
4th Jun 2008, 5:40 pm | #12 | |
Retired Dormant Member
Join Date: Jun 2008
Location: San Diego U.S.
Posts: 4
|
Re: Making a Triode Loftin White
Quote:
"C2 decouples the cathode directly to the steady HT supply voltage, which, as far as AC is concerned, is exactly the same as decoupling to ground" - This would be an accurate statement, if the power supply had low AC impedance to ground. But in that case you have to have two capacitors - the final power supply cap to ground, and the bypass cap to ground (or to B+). There is only one capacitor in the LW. In the LW, the inductor in the LW is not followed by a Cap to ground, so there is no low impedance path to ground. Instead, the AC current flows in a closed loop, that's floated above ground by the cathode resistor. In fact, you could replace both the power supply inductor and the cathode resistor with perfect constant current sources, so the AC impedance to ground is infinite, and the circuit still works. Now, the input grid is referenced to ground, but the output loop is not, as described above. As the output loop floats up and down due to power supply ripple, it generates a voltage potential between the input plate and output grid. The steady state power supply ripple is, of course, from imperfect filtering of the rectified AC. But there is also a dynamic ripple component that is caused by imperfect response to transient current demands, which will also cause the output stage to float up and down. The series RC, from the output cathode to the input cathode, injects the a portion of this ripple to input cathode, which generates an antiphase signal at the input anode, to cancel the ripple component. So this is not feedback as it does not compare the input waveform to the output waveform, it only subtracts the ripple component. It will not affect harmonic distortion or bandwidth or phase response. Pretty cool, no? Sheldon |
|
6th Jun 2008, 9:45 am | #13 | |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Darius Loftin White
Welcome Sheldon,
thank you very much for your excellent description. Quote:
for eliminating the influence of the output transformers primary-->secondary capacitance. It is the 4,5µF in the PC86 RL12T15 Loftin White schematic. Kind regards, Darius |
|
7th Jun 2008, 9:29 pm | #14 |
Retired Dormant Member
Join Date: Jun 2008
Location: San Diego U.S.
Posts: 4
|
Re: Making a Triode Loftin White
Thanks Darius. I overlooked that issue. How do you size that capacitor? By scope or calculation, then confirm? I also notice what looks to be a zobel across the primary of the OPT, but it's hard to read the values. Again, are those just done by looking at ringing and tuning accordingly?
Sheldon |
8th Jun 2008, 9:49 am | #15 | |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
Quote:
the 4µ5F was just there from the old circuit at the beginning of the blog. Do you mean the protection against arcing at the primary? Click on the link "Funkenstrecke" at the end of the blog. Kind regards, Darius http://rl12t15.blogspot.com/2007_09_01_archive.html (use a website translator, underlined words are hyper links.) |
|
8th Jun 2008, 9:10 pm | #16 |
Retired Dormant Member
Join Date: Jun 2008
Location: San Diego U.S.
Posts: 4
|
Re: Making a Triode Loftin White
Ah, spark gap. That symbol is new to me. Not something I've noticed on audio amp schematics before, though it seems like a good idea. Is it more necessary in a design like this, because of the cap in the current loop?
Sheldon |
9th Jun 2008, 8:29 am | #17 |
Retired Dormant Member
Join Date: Dec 2003
Location: Solingen, Germany
Posts: 727
|
Re: Making a Triode Loftin White
#16
Hi Sheldon, every output transformer should have some kind of protection against over voltage, shouldn't it? Kind regards, Darius |
9th Jun 2008, 3:57 pm | #18 |
Retired Dormant Member
Join Date: Jun 2008
Location: San Diego U.S.
Posts: 4
|
Re: Making a Triode Loftin White
|