For those who just can't get enough of listening test results on rectifier diodes (and also, in this case, transformer snubbing capacitors), I would point you to Quasimodo post #936.
Here is a link but it only works if you've set DIYA to view ten posts at a time; if you view fifty posts at a time, the link stops working. I wish I knew how to fix that.
listening results
A couple gems from that post:
Here is a link but it only works if you've set DIYA to view ten posts at a time; if you view fifty posts at a time, the link stops working. I wish I knew how to fix that.
listening results
A couple gems from that post:
- I buy very fast rectifiers [and] I get the highest current [rated ones]. The lower impedance of the high current rectifiers have deeper bass and the overall sound is more serious fuller and less like a toy. YOu still need to snub them. All parties in my group preferred the Vishay hyperfast welding rectifiers you can buy from digikey. They are like $50 for a block with two diodes in each block. Everyone in my group ponnied up for them. Some may find that they sound a little slow. I would say that are not hyped sounding like shoktys that sound faster than the music. Zippy. But there is far lower litening fatigue.
- In listening tests two .022 in series sounded better than one .01 uf. Everyone in the room preferred it. I suspect that this doubles the ESR and reduces the load. On the scope they looked the same. [this is "Cx" the snubber capacitor connected in parallel with the transformer secondary -- mj]
You have tried and found those diodes not to respond in anyway like that?
I'm glad to see that a number of people have independently tried different rectifier diodes and noted the differences. Like Richard Marsh has implied, it is the CONSENSUS of open listening opinions that shows the right direction, and the optimum diode is one of the factors that can make an audible difference. Take it from me, my best designs use fast-soft recovery diodes, have for decades.
A -1/4 power effect on noise (see Wurcer on this here and there). So higher Idss is better but not by much.
(Below) , they are modding an Adcom 565 with a "fancy bridge".
They hammer on about " AN INEXPENSIVE AND THRILLING IMPROVEMENT " ,
I have a vishay fast recovery test PS. My main amps , with any of the 6
classic input stages ... sound exactly the same with either the Vishay's
or a standard "poorman's" 25A bridge..
I ran a 45HZ signal to a sub , the "poormans" bridge (unsnubbered) exhibited
ringing ... but after the cap bank - nothing.
Even with this crappy setup , all sounded the same.
The only "audible difference" reflects a choice in a high PSRR
amp. The typical forum CFA seems not to have that tight bass definition
like a high rejection/NFB VFA.
An old 4 stage AB Sansui amp with cheap diodes + 10Kuf has more (bass) definition than so called hi-end OEM ??
OS
than some $3K
Attachments
Try taking the page number out of any links you create as they are settings specific. For yours it is the '-94'. If I create a link for that post I see '-47' because I have 20 posts per page.
So take that info out of any links.
Attachments
That's Id, i.e. gm goes as sqrt(Id) and noise goes as the sqrt(1/gm). If you pick a quiescent current for a particular device in the circuit it might be useful to have it be a certain amount less than Idss so Vgs does not go positive (N channel) under any dynamic conditions. This would be a reason to specify a higher Idss range. If you are going to run a 2SK170 at 1mA, there would be little reason to worry what range you pick.
Thank you. Good to see a respected member of this forum totally disagree. Let's not have any "you must be deaf or a liar".
Yeah I included the fast soft recovery diode used in your Vendetta Mk.II preamp (RURD460), in my comparison. Fairchild discontinued selling it in axial lead thru-hole packages, so I purchased and tested it in the SMD "D-pak" (TO-252).
Soft Recovery Diodes Lower Transformer Ringing by 10-20X | Linear Audio Volume 10
Your diode finished in the the 21st percentile (10th best out of 48). Acceptable but not spectacular. The #1 top finisher was from General Semiconductor, a company name which merged out of existence in 2001, so the diode itself is at least sixteen years old.
I would take the pos and neg wires from the terminal block, twist them around the two commons from the transformer from the plane of the terminal block all the way up to the splice, then twist each around the common to the terminal of the cap the white connects to, then onward to the terminal being connected to. That way, the uncommitted loop is the smallest it physically can be, just across the cap between the terminals. As is, you have a pair of broadcast loops.
John
Last edited:
With higher voltage Schottkys now available would they be a candidate for low level electronic rectifiers, or are the best selected fast / soft recovery diodes the better choice. (Yes i know test them and find out, but I don't have a spectrum analyser - I spent all my money on Beeswax fuses and now my system buzzes - another Eric the Half a Bee product)
- Status
- Not open for further replies.