|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
211.216.80.196
Hi Diyers,A few months ago I built above circuit and immediately found that the sound character is different from what I had experienced including all active parafeed, normal parafeed, normal OPT SE and PP. I could not describe the sound quality since it was different and new. Especially the bass is natural but the high freq was not so good. Later I got to know that parafeed DC blocking by cheap oil cap in the picture was the reason and I replaced it with Mundorf Tubecap 1000V 10uF from Ebay. Also replaced the coupling cap with Russian Teflon, now the high are all clean.
I've been listening to amplifier for more than a month and I'm so happy that I tried this circuit. It's in my office and all the friends including non-audiophile said it is good. Even connected to ProAc bookshelf speakers of low sensitivity the sound is good and balanced within the power. It is nice feeling that you can enjoy the orchestra music at low volume without low freq complaint. Next to this amplifier is Chimera-like DC coupled 300B SET, and phase split choke driven differential 300B PP and some Tripath digital amplifier. Pre amplifier is low DCR 5687 R load parafeed and Sowter 3575 - 12AX7- Sowter linestage PP OPT balanced linestage. I'm listening more with this R load 2A3 because the sound is close to 'right' and comfortable, not meaning somehow rolled off or compromised sound. Detail is there but with a feeling of comfort.
Steve Bench used the word 'magical' on his site and as a non-English speaker and novice DIYer I give up to describe further than Steve did. I also remember a local tube amp manufacturer saying simply 'very good.' on resistor loaded parafeed line stage.
I also tried autoformer connection but the difference is inaudible to my ears. I believe this is due to the high turn ratio (26:1) or the primary and secondary coil gauge is different. I am determined to try this with 300B or 300BXLS for more power with autoformer. By using fixed bias I can get rid of cathode R and C on the 300B.
I want to thank Steve for his site, JeffreyJ for encouragement and other AA inmates who mentioned of this and I recommend that tube DIYers try this for this new fun.
Cheers
C.Y
Follow Ups:
I've done a variation on this using 25 watt 240 volt light bulbs as the plate loads. It works out roughly 2000 ohms per bulb.What is interesting is that the bulbs change brightness levels as the music gets louder.
Like you, I think there is something different and good about the sound.
ray
Hi ray,
Very intriguing picture like the three faces looking at amp :-), it look like a pushpull or paralleled single? Change of the brightenss means the current throught the load is not constant I think. You know there's a transistor version of light bulb load by Nelson Pass. Either way I am curious of the sound.
It's a single ended parafeed design like yours with cheap Radio Shack line transformers wired as autoformers. The input tubes are 6C45PIs and the out put tubes are 829Bs wired as triodes.I taken it to a few of the meets that we hold a few times a year in the Philadelphia / Washington area and it usually draws good comments. It's easily the best sounding thing I have built and it is the clarity which is a step above any other amps I have tried to build. There may be other causes of this since this was my first amp with extensive use of motor run caps in the power supply and of course umpteen other variables.
I had read the Nelson Pass article and had thought about trying to build it, but I was heavily into tubes by then. But I must admit his article was the inspiration. I had also read Stve Bench's stuff on the resistor load and that was part of the inspiration as well.
I've thought about it for quite a while now and I still have no idea why resistor loading might be a good idea.
ray
Hi ray,Today I replaced the parafeed cap and the soundis getting better. Yes this is also the best sounding amp I have ever built. The clarity shines across the audio band and it makes me stick to the listening position.
If possible could you tell me more on the line transformer that use for autoformer? I want to try this circuit with 300B and am at a loss what to use. I look at the other pictures of the library gathering and it must have been a pleasant meeting I guess.
Radio Shack was selling these 70 volt 10 watt line transformers until about a year ago. They were well under $10 each.The input side of the transformer has taps for 10, 5, 2.5, 1.25 and .625 watts. There is also a tap marked C.
The output (speaker) side has taps for 16, 8, 4 ohms and a tap marked C.
Used as an ordinary parafeed transformer and depending on the turns ratio required, I would connect the output capacitor to the .625 tap and the C tap to ground. Then the speaker to the appropriate taps on the output side.
As an autoformer, instead of connecting the C tap on the input side to ground, connect it to 16, 8, or 4 tap that you are using to connect to the speaker. The C tap on the output side is connected to ground.
Of course for safety reasons, it is essential that the output capacitor be situated between B+ and the transfomer. Thse are cheap transformers and I would not trust them with the cap between the transformer and ground.
This autoformer thing is easy to try. You may like it or maybe not.
Our group has had a wonderful time with our meetings. It really has encouraged attendees to start building, share ideas and be prepared to experiment with different ideas. There has been some wild stuff appear over time. Even the failures are interesting. If you don't belong to such a group, I would encourage you to form one.
Thank you for the info. I do not have any place to source those line tranformers but have one local winder who can do anything for me, though the result is all my responsibility. What I know now is that the coil gauge should be somewhat close to the secondary of normal OPT, resulting in a big size like yours.I'd love to join your group if possible. Tube DIYers in my country are normally obsessed with Marantz 7 ciruit or something that is already 'proven'.
Maybe resistor loaded autoformer parafeed will benefit more with DC coupled circuit, which is in my mind at the moment.
It would be interesting to see how it compares to choke loaded or CCS loaded stage of the same topology. Resistor loading is kind of wasteful, so what is the gain for the loss of power?
I agree that a comparision would be interesting.To answer your question on the advantages, .. check out the section "Resistor Load, Capacitively Coupled to the Output Transformer" in the page C.Y. linked to.
Hi,I have not done the direct comparison of CCS, Choke,and resistor load in a same cicuit. But I have tried many different kind of CCS inclluding C4S, many by Gary Pimms, DN2540 and also different plate chokes. My impression on CCS is that the net result depends on the componets before and after the circuit you are using. For example if you are using so so linestage and use all active loaed amplifier you end up revealing the so so linestage's character, including the bad one. Active load is better used close to the source like tube DAC. On the other hand plate choke is very much like a 'person' and the character is all different. Maybe because there are many factors in making plate choke I think. But I do believe a well made plate choke would give me nice result. But you know we are simply enjoying this hobby, not pursuing a 'ultimate winner'. Also I do not mind using a little more power for fun of tube DIY.
What I want to share is this charming experience with resistor loaded parafeed, and it's easy or even inexpensive than using plate choke. Unlike plate choke a resistor character is close to plain or honest I think. Unfortunately I am not good enough to perform a measurement here.
Seems like the big dropping resistor would cut output a lot, but would also smooth out the load seen by the output anode, and make a "reactive" load more "resistive". Thanks for reporting the experiment... --keto
This post is made possible by the generous support of people like you and our sponsors: