Bob Cordell said:
Mr C,
saving space may be relative, considering that Jensen 4-poles have the screw terminals on both ends of the capacitor.
I think i speak for a number by stating that i got the willies from endless variations of the Hitachi application note, to get more out of the lateral MOSFETs.
Sort of makes me feel relieved that another one of you audio senators, Mr John Curl, gave up too.
The collectors bug has been bothering me again lately, today received some more Sony VFETs and Hitachi 2SJ50/2SK135s from the mailman. ($2/pc. D..., i'm so good i feel like scavenging the beach tonight)
A few more of the same class VFETs, and i have enough to rebuild the Marantz class A MA-24 schematic for depletion mode devices. For a circuit/parts like that the Jensen caps might be a nice addition.
4-pole capacitors may be easier to integrate into a design than how i put it, but i still think the entire thing should be prime beef quality for them to be worthwhile.
I've been noticing Jensen 4-pole caps being used for just about any circuit lately, just as every salami is being stuffed with Rikens, Caddocks and Blackgate material.
Attachments
Mr Curl,
there's an electronics store in France that sells Sony V-FETs, but the ticket is $25/pc.
I'm finding them by the piece or in small batches everywhere, both used and NOS. Downside of course is that it takes a lot of patience to collect enough of the same Vp type.
Buying new is easier, i took your advice and roamed the globe for dual JFETs. Bought them by the 100s and hijacked most of them from your parts. Equal numbers of 2SJ and 2SK types, both BL and V types.
Cost me $1.15 on average, did not buy a single one in Europe, i may have more than you.
Mr Cordell,
about 20 years ago i made a phonecall to Germany to talk to the guy below, it's a 1987 picture.
I asked him for his advice on rebuilding his 100 watt/8 class AB design that used RCA RFP Vertical FET output devices to 50 watts in class A.
To my surprise, a year later his audio company put a 50 watt class A version on the market.
Old story, but still funny imo. I've constructed NP's A75 design with IRF630/9630s, also built Norman Thagard's A100 class A amplifier which has stacked Vertical FETs and Motorola BJTs in the output stage.
I see audio as cooking, you can really enjoy eating truffles but plain scrambled eggs with cream tastes just as fine in the morning.
I know your flavor, if the CIA had the desire to keep an eye on you audio guys they'd be hiring me.
The audio company : AVM
there's an electronics store in France that sells Sony V-FETs, but the ticket is $25/pc.
I'm finding them by the piece or in small batches everywhere, both used and NOS. Downside of course is that it takes a lot of patience to collect enough of the same Vp type.
Buying new is easier, i took your advice and roamed the globe for dual JFETs. Bought them by the 100s and hijacked most of them from your parts. Equal numbers of 2SJ and 2SK types, both BL and V types.
Cost me $1.15 on average, did not buy a single one in Europe, i may have more than you.
Mr Cordell,
about 20 years ago i made a phonecall to Germany to talk to the guy below, it's a 1987 picture.
I asked him for his advice on rebuilding his 100 watt/8 class AB design that used RCA RFP Vertical FET output devices to 50 watts in class A.
To my surprise, a year later his audio company put a 50 watt class A version on the market.
Old story, but still funny imo. I've constructed NP's A75 design with IRF630/9630s, also built Norman Thagard's A100 class A amplifier which has stacked Vertical FETs and Motorola BJTs in the output stage.
I see audio as cooking, you can really enjoy eating truffles but plain scrambled eggs with cream tastes just as fine in the morning.
I know your flavor, if the CIA had the desire to keep an eye on you audio guys they'd be hiring me.
The audio company : AVM
Attachments
John,
> where did you find Sony vfets?
I used to have a source who had some 20 odd Sony VFETs (2SK60) of the same ranking, and he was kind enough to send me the lot for measurement.
Apparently, after years of ageing, they are not that well matched anymore (both in Vgs and in transcondyuctance). In the end, I returned the lot and just bought one as a museum piece.
I then got some Loveltech JFETs from Grey Rollins and measured them together with a cascode (2SK1529). They are, IMHO, just as good if not better in every aspect. The only penalty is that you burn a few volts because of the cascode. But there are enough supply of devices around and one can also change the cascode device to change the load line, (almost) independent of speaker impedance. So I would not hesistate to recommend to anyone to consider that approach. Thanks Nelson for the tipp.
http://www.diyaudio.com/forums/showthread.php?postid=1089432#post1089432
Patrick
> where did you find Sony vfets?
I used to have a source who had some 20 odd Sony VFETs (2SK60) of the same ranking, and he was kind enough to send me the lot for measurement.
Apparently, after years of ageing, they are not that well matched anymore (both in Vgs and in transcondyuctance). In the end, I returned the lot and just bought one as a museum piece.
I then got some Loveltech JFETs from Grey Rollins and measured them together with a cascode (2SK1529). They are, IMHO, just as good if not better in every aspect. The only penalty is that you burn a few volts because of the cascode. But there are enough supply of devices around and one can also change the cascode device to change the load line, (almost) independent of speaker impedance. So I would not hesistate to recommend to anyone to consider that approach. Thanks Nelson for the tipp.
http://www.diyaudio.com/forums/showthread.php?postid=1089432#post1089432
Patrick
Re: Power supply quality
Dear Mr Cordell,
thank you very much for your very interesting post.
Honestly I scrolled the best part of the 3D but I could not find the answer to your very interesting initial questions I quote.
After all is there a consensus that the quality of power supply is more important for the first gain/driver stages?
All this makes a lot of sense to me.
Moreover, I wonder if a switching mode power supply, switching outside the audio bandwidth, is the best way to power these first and very "sensitive" first stages.
I am very interested in this discussion beacuse I am going to use a switching power supply in a line preamp, convinced that this is the best solution, after all.
Thank you very much indeed.
My kindest regards,
beppe
Dear Mr Cordell,
thank you very much for your very interesting post.
Honestly I scrolled the best part of the 3D but I could not find the answer to your very interesting initial questions I quote.
After all is there a consensus that the quality of power supply is more important for the first gain/driver stages?
All this makes a lot of sense to me.
Moreover, I wonder if a switching mode power supply, switching outside the audio bandwidth, is the best way to power these first and very "sensitive" first stages.
I am very interested in this discussion beacuse I am going to use a switching power supply in a line preamp, convinced that this is the best solution, after all.
Thank you very much indeed.
My kindest regards,
beppe
Re: Re: Power supply quality
I don't think a switching power supply is helpful or necessary for sensitive stages. The fact that its switching frequency is beyond the audio band does not necessarily help. Moreover, extremely clean linear supplies can be made for circuitry that has low power consumption.
Best regards,
Bob
beppe61 said:
I don't think a switching power supply is helpful or necessary for sensitive stages. The fact that its switching frequency is beyond the audio band does not necessarily help. Moreover, extremely clean linear supplies can be made for circuitry that has low power consumption.
Best regards,
Bob
MikeBettinger said:
I think that linear power supplies are the power supply of choice for low-current devices like pre-amps because efficiency there is not such a big deal, and because noise susceptibility in these circuits to HF noise radiated by a switcher is probably more difficult to tame than 60/120 Hz noise and rectifier switching in a linear supply. But this is only speculation.
In power amps, the situation may be a bit more complicated. Some big power amps are now using switchers, like the Halcros. I'm not sure whether their use really does anything for the sonics, though. I think a key thing when using switchers in a power amp is that one must not be lulled into using a smaller reservoir capacitor than one would use if the PS was a linear one. The transient current reserve available from a 22,000 or more uF reservoir cap per rail is important. If one uses a switcher in a power amplifier, and gets past the EMI issues, there are then two likely advantages. The first is the absence of 120 Hz ripple on the rails, especially under heavy load conditions. The second is that regulated rails largely come for free as part of the switching supply deal. The downside of this, of course, would then be the absence of dynamic headroom in the amplifier as a result of the traditional compliance of a linear supply. Here we get into the individual designers' philosophy.
Bob
beppe61 said:
If you are talking about the amount of reservoir capacitance in a power supply, generally, the more the better. Yes, it will "slow down" the rise of the power supply voltage at turn-on, but that is not an issue for sonics. If you use a lot of capacitance with a fairly beefy transformer, your only concern may be that you will need to incorporate a soft-start (inrush control) circuit to avoid blowing a circuit breaker or over-stressing your power switch contacts.
For a power amplifier, 10,000 uF per rail is fairly skimpy, while 100,000 uF per rail is nice if you can afford it.
Bob
MRupp said:
except you will never achieve this ideal being voltage will not change regardless of load current. The only exception that I can think off is that you live in Arizona and the stereo is hooked up to some heavy duty solar panels and one only listen to music when the sun is bright and no cloud within miles.
I assume that there will be "some" voltage fluctuation based on current consumption. Consequently, I concern about the time required to recharge the capacitors. Maybe I just like the sound of underdamped PS. I just don't think a blanket statement that the more capacitance the better always is valid.
I like to add that my experience with solid state PS design is limited to CLC filters, as I always use filter chokes. So, it may be different for those that only use a large C as the only means of filtering.
I am working on one right now that is CLC. Each power rail of each mono block has
22000uf --> 40mH 16A choke --> 68000uf
So it does have quite a bit of capacitance. And PSUD2 indicated that I may get better result if I lowered the capacitance of the last cap.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.