• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

GB NP matched Toshiba 2SK1530/2SK201 for F5X

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
EUVL

EUVL

Member
Joined 2003
xen-audio.com
No. I am also waiting like all of you.

Smyslow is the only person who has both amps (K1530/J201 & K3497/618) within the test team.
But he is too busy right now to have a chance for proper listening.

I expect the two sets of output devices to sound different.
Measurement does not tell you everything.


Patrick
 
buzzforb

buzzforb

Member
Joined 2010
EUVL said:
No. I am also waiting like all of you.

Smyslow is the only person who has both amps (K1530/J201 & K3497/618) within the test team.
But he is too busy right now to have a chance for proper listening.

I expect the two sets of output devices to sound different.
Measurement does not tell you everything.


Patrick
Click to expand...

I would tend to agree, as every pair I have tried in the Ba output have sounded different, but some more than others. The laterals were the most different in terms of sound, others more nuanced. If comparing to set of fets that make for a pretty good complimentary pair like 3497/618 vs 1530/201, how differerent will they be. Guess i should plug the 3497/618 into the BA and listen for differences. Guess timw will tell.

FLG:cuss:
 
P

pchw

Member
Joined 2006
The cut & paste sometimes have this kind of issue that both users can start with the same list; hence, the last save overwrites the first post. Use the Google doc (in post #1) which has a little concurrency control and reflects any other's changes while you are editing.
 
EUVL

EUVL

Member
Joined 2003
xen-audio.com
> 3497/618 are not true complementary. 1530/201 are.

That is correct.
But then 214/77 or 2013/313 are also not true complementary.

3497/618 has high transconductance but also high capacitance.
It is like the discussion a few years back of IRFP044 vs IRFP240 in Aleph X.
The former sounds darker.

I expect 3497/618 to be similar.


Patrick
 
P

pchw

Member
Joined 2006
Hee hee, maybe it is just me.
The first "official" F5X amp will not be the experimental platform, no matter what people will say about the sound of different parts. Various exploration belongs to the second build using the diy store's soon to be available chassis :)
 
N

Nikon1975

Member
Joined 2009
I am traveling now, so not quick to respond.
Basically to have 500+500 pieces, we have to spend 5000 euros upfront, roughly.
I am leaving Japan end of June, so this should be decided quickly.
Good part is that I could carry the devices to Europe or even to USA.

I would be also interested in two sets.

D.
 
A

AndrewT

R.I.P.
Joined 2004
If we can get a yield of "good enough" pairs of about 50%, then based on interest so far then a Group Buy of 500N & 500P should be about right.

We can sell off the unmatched 50% @ a 25% discount on our Group Buy purchase cost to recoup quite a bit of our expense. Those unmatched devices could be grouped as single point pairs to bring up the total yield to 99.6%
That way everyone gets good value.

I would certainly be interested in buying a stock of single point pairs for future projects.
 
P

pchw

Member
Joined 2006
I am sorry to disappoint the group. As I mentioned twice, I will not be the organizer of the GB, or I should say I won't have time to participate in the next 2 months (unless people can wait:)). I will be traveling soon.

I really wanted to help this taking off; hence, I created the thread and the spreadsheet. I hope someone will lead to move forward. I even gave this a very good thought of the logistics. Here are MHO:
  • Decide the minimal Vgs that is considered as matched, e.g. 1%, 2%, 10mV, 20mv, 50mV.
  • Since everyone wants to the best match, it will make sense to set up a 3 tier pricing for match sets. Let say, the minimal acceptable match Vgs is 100mV, then there can be 3 pricing for 3 different ranges.
  • As of now, there are 42 sets in the spreadsheet. Since there is only one non NP mach sets, I will estimate as if all are NP match. This will mean 168 matched NP pairs, and majority are 4 match NP pairs. 168/500 implies 33.6% match rate. However, it is important to get a confirmation for accurate count. I added a confirm column to the spreadsheet for this purpose.
  • Rough calculation. Assuming 40% yield to meet the minimum Vgs, then there will be enough to provide NP match sets to all. 5000 Euros (or USD 6610) for the devices, 5000 Euros / 42 sets = 120 Euro (USD160) which will be the base line of a matched F5X set. **It is important to realize that this is a bottom up estimate. The actually cost will be more once you add the overheads (exchange, paypal, shipping etc, expect at least 10% more) . But it is a starting point to estimate how much fund to collect from participants to purchase the devices.**
  • If the yield is perfect, then no tier pricing is needed. If the yield is uneven, then the price will be adjusted according to the match tightness. Additional charge for tighter match will be donated to the forum for a good cause.
  • AndrewT has an excellent point about the "left over". I suggest to distribute the non NP mosfets in proportion to the number of sets purchased to the participants. As it stands, each match set will have additional 14-15 devices. These are likely NN and PP matches as well!!!
  • Buzzforb offered to help matching per Patrick's 3 point matching in F5X operating condition (thank you). No free labor, so there should be an additional cost added on the top of the devices for this effort.
  • Total cost then will be devices + Buzz's matching work. I hope Buzzforb will accept a couple free F5X match sets (I know you want some anyway) as the reward instead of additional $$. This way there will be no additional cost to the participants. As from above rough estimate, 2 matched sets will still be equivalent to USD360+.
  • Assume Buzz will accept the non-monetary pay, and there are matched F5X sets left. I suggest to send Patrick and Mark Lai each one set as an appreciation gesture. If there are still more left, we will figure out. Maybe donate to the forum??
  • Due to the unpredictable nature of matching devices. I suggest to close the GB once it reaches 60 confirmed F5X sets.
  • Last not least, come up with a "what if the yield is crappy" plan, i.e. there is not enough F5X sets to meet the group. This will be important to avoid future dispute.

I hope this will provide enough detail to move the GB into another stage.

Have a great weekend!!
 
EUVL

EUVL

Member
Joined 2003
xen-audio.com
> Assuming 40% yield to meet the minimum Vgs

I would consider this optimistic, though I have no data to proof.
I think you should aim for a 10mV NN PP (or NNNN PPPP) match, and maybe 50mV N-P match.
This I consider more realistic.
And unless you have great equipment, I do not believe in better than 10mV match.
(due to device thermal drift during test)

It is a great gesture that you would offer us a set or two.
But our aim is that you can build great amps with our "assistance", and we have enough device in reserve.

We might wish to buy some of your "reject" devices at discounted price though. :)

Someone has to come forward to take the lead.
And we shall not.
Else we shall undoubtedly charge you for our service, which is not in your interest.


Best,
Patrick




Patrick
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom