Excellent experiment! Thanks for sharing this. I have PIO, FT-3 and some iCW, so will try them!
Cheers,
Ale
William,
Thank you for sharing your tests! Very informative and much fun to read. One thing that bothers me is that in all tests the Oscons were in. Have you tried to disconnect these on one side and then try lager values (10u or 4.7u) of foil caps instead? It could be that the presence of an electrolytic cap blocks in a way the qualities of the foil caps you tried.
(Could be you tried that long time ago before the Oscons went in.)
Peter
Thank you for sharing your tests! Very informative and much fun to read. One thing that bothers me is that in all tests the Oscons were in. Have you tried to disconnect these on one side and then try lager values (10u or 4.7u) of foil caps instead? It could be that the presence of an electrolytic cap blocks in a way the qualities of the foil caps you tried.
(Could be you tried that long time ago before the Oscons went in.)
Peter
I tried a lot of foil caps before trying the oscons, and the oscons were best at the time as an unbypassed bypass for the LED's. The pre had been sounding a bit thin and the 330uF brought back the missing "body" to the sound. I can't guarantee that I tried ALL of them though. I have never been too focussed on this pre as it is still a breadboarded experimental amp. I like what I am hearing now with this extra bypass, but the microphonics are still hectic. I can sort out a proper top-plate and mount it properly to the chassis, but I am not convinced that will be sufficient. The valves ring if the dogs bark, or if you clap your hands. Anyone got any brilliant ideas for quieting them (that don't involve days of work or large sums of money or look like a dog's breakfast)
The polymer caps and other super-low ESR caps are available from computer modification enthusiasts, such as Badcaps.net. Those guys are awesome! I got a range of different makes and types to try for DAC's. They have Oscon polymers and Nichicon polymers as well as loads of other super-low ESR makes and series. Values tend to be rather large and voltages low, because they are mainly for motherboards, but there is no minimum order quantity and prices are silly compared to other major suppliers. (Hope this post doesn't start a run and push prices up!)
I think I will try some IR led's in the future and grid biasing is definitely on the cards for my 4p1l power amp. I did use battery biasing in this pre to begin with, but just got the hell-in with those 9v batteries dangling all over the show whenever I wanted to modify something. I don't recall that it sounded any better than LED biasing (WITH bypassing that is).
The polymer caps and other super-low ESR caps are available from computer modification enthusiasts, such as Badcaps.net. Those guys are awesome! I got a range of different makes and types to try for DAC's. They have Oscon polymers and Nichicon polymers as well as loads of other super-low ESR makes and series. Values tend to be rather large and voltages low, because they are mainly for motherboards, but there is no minimum order quantity and prices are silly compared to other major suppliers. (Hope this post doesn't start a run and push prices up!)
I think I will try some IR led's in the future and grid biasing is definitely on the cards for my 4p1l power amp. I did use battery biasing in this pre to begin with, but just got the hell-in with those 9v batteries dangling all over the show whenever I wanted to modify something. I don't recall that it sounded any better than LED biasing (WITH bypassing that is).
Peter, I may still try larger value foil caps on their own at a later date, though I don't have any good quality 10uF foil caps at the moment. I think I have some 3,3 ICW's. I have some motor-runs, but they are not in the same league quality-wise. Unfortunately, the Oscons are soldered very neatly into a board I made up for the LED's and I really don't want to desolder them and then try top put them back again, as this will no doubt damage the board.
Mogliaa, I would like to hear what your results are with only largish foil caps,though I would also encourage you to get some polymer oscons and compare them. (See post above for a good supplier).
Mogliaa, I would like to hear what your results are with only largish foil caps,though I would also encourage you to get some polymer oscons and compare them. (See post above for a good supplier).
Oh, BTW, I noticed while doing all of this that when the LED's got really close to a valve, there was an audible hiss through the speakers. Worth thinking about?
Does anyone have any 26's to spare? I would dearly love to hear for myself what that much-vaunted DHT sounds like and compare them to the 4p1l's.
Does anyone have any 26's to spare? I would dearly love to hear for myself what that much-vaunted DHT sounds like and compare them to the 4p1l's.
Last edited:
OS-CONs are one of the few industrial components that actually discussed audio applications in detail. Or rather the 1990s version of the documentation did (Sanyo OS-CON Technical Book). The latest version has that all cut out, huh.
Anyway, there are precautions to be aware of when using them in audio:
- Take care to protect them from mechanical shocks - even boards knocking into each other is noted in the manual as injurious. Make sure your supplier ships them with adequate protection, and not just in a bag.
- There is no transient overvoltage spec (surge voltage) like normal electrolytics. If they see even a brief overvoltage, thet fail short circuit. I recommend use of a zener if you apply them in a cathode circuit.
- watch out for the leakage in the polymer versions SVP/SEP. For a 16V 470uF SEP for example, the leakage is 1.5mA (!) This current is not stable either, and can play havoc with your biasing, especially in non-LED cathode circuits. The leakage problem means they should not be stacked in series, either. The old SA and SC versions are better than this, but watch out nonetheless.
- The old manual even quoted a break-in time for audio use (don't listen for 10 hours, and expect improvement all the way up to 100 hours)
- they also recommended to leave the cap in place one soldered - unsoldering annoys the audio-pixies inside.
-
Anyway, there are precautions to be aware of when using them in audio:
- Take care to protect them from mechanical shocks - even boards knocking into each other is noted in the manual as injurious. Make sure your supplier ships them with adequate protection, and not just in a bag.
- There is no transient overvoltage spec (surge voltage) like normal electrolytics. If they see even a brief overvoltage, thet fail short circuit. I recommend use of a zener if you apply them in a cathode circuit.
- watch out for the leakage in the polymer versions SVP/SEP. For a 16V 470uF SEP for example, the leakage is 1.5mA (!) This current is not stable either, and can play havoc with your biasing, especially in non-LED cathode circuits. The leakage problem means they should not be stacked in series, either. The old SA and SC versions are better than this, but watch out nonetheless.
- The old manual even quoted a break-in time for audio use (don't listen for 10 hours, and expect improvement all the way up to 100 hours)
- they also recommended to leave the cap in place one soldered - unsoldering annoys the audio-pixies inside.
-
Thanks for the detailed information Rod. I have used Oscons for 12 years and have experienced 1 blown one (It was my fault, not the cap's) Perhaps the New polymer caps are different - I don't know. I received mine in a bag with a whole bunch of other caps and they all work fine. The leakage current is a bit of a worry for some applications I guess, but they seem to work fine here. Leakage current for mine is shown as 792uA max. They are rated for ripple current of 4.7A. I suspect that with the kind of mA currents in most preamp applications, leakage current will be a whole lot lower, but am open to any knowledgeable input on the subject. BTW, the ESR rating is measured at frequencies of 100 to 300Khz. It will be a whole lot lower at audio frequencies, will it not?
I am sorry you told me about the audio pixies.
That is something I never had to worry about before; now I need to find out how to keep them happy. Do you think maybe if I placed some quartz crystals in the amp.....? No doubt they are the secret ingredient in bybee filters. Those things must be filled to the brim with them!
Anyone know of a dealer who sells pixies loose?
I am sorry you told me about the audio pixies.
That is something I never had to worry about before; now I need to find out how to keep them happy. Do you think maybe if I placed some quartz crystals in the amp.....? No doubt they are the secret ingredient in bybee filters. Those things must be filled to the brim with them! Anyone know of a dealer who sells pixies loose?
Yes! My test were only with CCS although were plainly frequency and THD response as had only one driver and didnt connect to amplifier for a proper listening test - which I'm planing to do shortly
Cheers
Ale
OK. I will wait to hear your opinion on CCS loading with this tube. My experience on 6n6p and 6cg7 with gyrator and CCS loading has not been promising.
I have to say that bypassing the LED bypass cap has definitely paid off in a big way. PIO rolled off a bit of HF, but I now have hooked up K75's 0.22 across them and it is very well balanced and very clear indeed.
I have to say that bypassing the LED bypass cap has definitely paid off in a big way. PIO rolled off a bit of HF, but I now have hooked up K75's 0.22 across them and it is very well balanced and very clear indeed.
What was you didn't like about both options?
Had order some OScons to add to the 100nf PIO I have at hand.....
Cheers,
Ale
Hard aggressive sound and audible hissing from tubes with these loads. I am thinking of trying a sort of "hybrid" load - bypassing a gyrator or CCS with a large value resistor to try to get the best of both worlds. Lowish drop in B+ compared to a large resistor; more linearity and better gain than just a resistor; less hardness than only a CCS. Tunable sound by varying the resistance according to taste.
If you are going to install 330uF Oscons, you will probably need to bypass them with something bigger than 100nF to get a maximal audible improvement. My PIO's are 3,75uF.
If you are going to install 330uF Oscons, you will probably need to bypass them with something bigger than 100nF to get a maximal audible improvement. My PIO's are 3,75uF.
Here is where I am. I'm in the process of building a modular chassis for a preamp so I can swap different valves. I will have three top plates. One for the preamp connectors and filament boards, second one for the OT or bias board and third plate for the valves. Therefore I can easily try different connections.
My first test will be 26 with LL1660 and filament bias.
Btw: Got the OSCONs delivered yesterday...
Cheers,
Ale
I also ordered some Oscons after reading Hihopes positive results. I tried them a few days ago to bypass cathode leds that had already foil caps as bypass. I must say that it sounded pleasant and natural with the Oscons installed. But after removing them and listening to foil cap bypass only there was more air around instruments and everything was more lively. It was also easier to hear sounds decay.
Peter
Peter