sic NPN (SAND) upgrade for D1B1??

Status
Not open for further replies.
Jump to Latest
qusp

qusp

Member
Joined 2009
#1
Hi Guys, i'm a SS fan, but I do have to admit to knowing very little about possible mosfet upgrades and what effect if any they have on sound quality. I can understand why they would help to make an amp more efficient or switch faster and at higher current, but arent we already past what we need there? i'm not so much skeptical, just curious and not at all beyond tweaks some might call dubious, so thats not my angle

now if I havent scared you off with the above question, I want to get some advice on how to test this for myself, im building the D1B1 symmetric buffer for my sabre dac (ackodac) and I would like to get some suggestions about how to best burn a couple hundred dollars on 8 fets (or less if anyone has any great suggestions), possibly even more like 40 a piece on the ones driving the cables and only 10-15 a piece for the other 4. the schematic calls for IRF610's, all are NPN for those not familiar with the schematic. I tried to PM Papa to see if he knew how I could upgrade the sand, but I guess due to overwhelming demand for his time, he has had to chose not to accept PMs, (which I totally understand) so thought i'd ask you guys.

I figure although this is a fairly low current application, I know the sabre sound very well and since there are only a few active devices in this design it might be a good opportunity to get my head around the 'sound' of different fets.

if you mods feel this thread might be best in the parts forum I apologize, feel free to move it, but since its fairly specific to this design at the moment I thought it might be a good place to ask. plus of course there is the possibility of Mr Pass himself popping his head in. thanks for putting the original design out there Nelson 😎, I look forward to getting it running on all cylinders

thanks for reading
 
#2
oh I should probably mention I have been looking at the sic devices from IXY and some schottkey from cree and IXY for the power supply, but wondered how this performance increase for power conversion translates into the audio band. also read good things about GaN, but same story
 
#3
I have been playing with a number of new exotic parts, and
I can tell you that the interesting thing about them is not
the speed or low Rds, rather it is the transfer curves as
reflected in the distortion characteristic.

Some of these parts do not represent an improvement in this
regard, but a few exhibit much lower distortion. So far, the
lowest figures are about 20 dB better than the IRFP240.

😎
 
#4
aha ICIC and am I on track with the manufacturers above Papa? IXY in particular? come on gimme some sugar!!..ermm sand. I already enjoy the IRFP240, so those numbers are quite exciting to me, especially when trying to match the 140db DNR of the sabre. very excellent! now to go check out some transfer curves. thanks for the heads-up mate!!
 
#5
Dear Nelson,

Have you tried Semisouth R550 in that position?
I'm interested to know how will it perform in the place of IRF610 in D1, but lacks the proper setup and equipment to do the measurement.
If you have interest to test R550 and have yet to do so, I can sent you a pair or a quad of R550.

Regards,
Zhou Fang
 
#6
I spent some time with the R550 last weekend. As an output
device it has more bandwidth than the other power parts such
as the R100's or IRFP240's, and this is due to lower capacitance.
With that comes greater distortion at lower frequencies. For
this it would be natural if your source impedance was high or
you were building a tweeter amp.

The IRF610 still has lower capacitance, but also much lower
transconductance, about 1/5 as much.

I don't know what a D1B1 is, but if it's the IV for a DAC, then
you are better off with JFETs like the 2SK170's, maybe a
few in parallel.

😎
 
#9
Nelson Pass said:
I spent some time with the R550 last weekend. As an output
device it has more bandwidth than the other power parts such
as the R100's or IRFP240's, and this is due to lower capacitance.
With that comes greater distortion at lower frequencies. For
this it would be natural if your source impedance was high or
you were building a tweeter amp.

The IRF610 still has lower capacitance, but also much lower
transconductance, about 1/5 as much.

I don't know what a D1B1 is, but if it's the IV for a DAC, then
you are better off with JFETs like the 2SK170's, maybe a
few in parallel.

😎
Click to expand...

Hi Nelson, thanks yes it is indeed the D1 I/V but amputated and used as I/V and buffer for any current out dac. the circuit was taken from the D1 and there is soon to be an SMD version as well. check the latest revision by owen which is SMD and omits the buffer, but I have the PCB pictured on this page

thanks for the heads up though, I will grab some 2SK170 as well to try
 
#10
yes the problem with them is heat though, I think I will grab some, but not sure if it will work with this PCB very well. this PCB is pretty much exactly the D1 I/V and buffer, which I am using until either Owen puts out a PCB or I get off my *** and do it on perf. as you know it uses 610's is it just learning that makes you choose Jfets here now or something else?
 
#11
Funny how I couldn't interest any of the DAC players in that
circuit back when there was a lot of business in that category.

One major company looked at it and said they didn't see how it
could possibly work. I said, "Are you saying that my sample
sounds bad?"

😎
 
#16
Hi Nelson, yeah well its actually its simplicity that appeals to me, less components in the signal path to me = better sound in most cases. I will be trying omitting the buffer stage as well like Owens. decided to go ahead using IRF540 for the moment because I have some here and the PCBs arrived yesterday 🙂 but have ordered some other parts to try out as time goes by. including some 2SK170 and getting some sic devices from IXY in this digikey order, undecided which one at the moment. I like simplicity so much that often I like passive I/V for dacs with transformers or R's, but the sabre performs so much better into a low impedance node that a SS circuit is really the only option
 
Last edited:
#17
EUVL, I would be interested in your version without caps, I was intendding to try to achieve the same thing with my build at some point(easier because my whole signal path is balanced) I assume there is a base level of device matching that must be maintained to avoid offset, or have you used a servo of some sort?
 
#19
I already am and that was the idea. i've built up 4 different output stages for the ackodac so far for testing and will setle on the preferred one, but I always like the idea of less caps in the signal path so yours appeals as another option. some are simple (transformer coupling and the D1 I/V ) and some like peranders QRV08 (2 of them for dual mono) and perhaps Erno's EB937 are more complex
 
Status
Not open for further replies.
Top Bottom