DIY Class A/B Amp The "Wolverine" build thread

Hi guys, I see that apparently the only difference between 2SC5200-N and 2SC5200-O is h_fe min: 55 for the former and 80 for the latter (whereas h_fe max is the same). Can the two be used interchangeably or has h_fe min a specific role?
No, use only the O rating. ONSEMI 2SA1943OTH and 2SC5200OTU are good, they have TO-264-3 package. Mouser numbers 512-2SA1943OTU and 512-2SC5200OTU
 
@mainframe99 From the Toshiba 2SC5200N datasheet
https://eu.mouser.com/datasheet/2/408/2SC5200N_datasheet_en_20150512-1799632.pdf

Using continuously under heavy loads (e.g. the application of high temperature/current/voltage and the
significant change in temperature, etc.) may cause this product to decrease in the reliability significantly even
if the operating conditions (i.e. operating temperature/current/voltage, etc.) are within the absolute maximum
ratings. Please design the appropriate reliability upon reviewing the Toshiba Semiconductor Reliability Handbook
("Handling Precautions"/"Derating Concept and Methods") and individual reliability data (i.e. reliability test
report and estimated failure rate, etc).

Ensure that the junction temperature does not exceed 150.
 
  • Like
Reactions: Mainframe
Hi guys, in order to come up with a matched pair of transistors how many of them do I need to buy?

The problem is that if I need 6 matched pairs, I have to buy a LOT of transistors and it seems a huge waste of money to me...correct me if I'm wrong

And even then, what do I do with the remaining transistors?🙁
 
Hi everyone.

Do we really REALLY need to go through transistor matching for output devices and drivers? I'm asking because I made some cost estimates and it really makes the cost of the project skyrocket. At 2.50$ per device (minimum!), say I need to buy 30 PNP output devices and 30 NPN devices to come up with 4 matched pairs (and I think I'm being tight). It makes a total of 150$ PER CHANNEL. It means a total of 300$ only for output devices.
Then, we have to match drivers.
Then we need to shell out additional money to buy a suitable matching rig.
And in the end I'll be left with a lot of transistors which I know I'll never use again.

Do we really need to splurge 500$ or (very likely) more just for transistor matching? I didn't see anyone mentioning this cost in this thread...

If I could just buy the required number of transistors and be happy with it I would really appreciate it.
 
Thanks Stuart. Comparing the THD measured using unmatched transistors (reported in the post you addressed me to) to the THD reported in the build guide, is it correct the say that the use of matched pairs will reduce THD by a factor of 5 to 7 with respect to a unit with unmatched pairs?
 
is it correct the say that the use of matched pairs will reduce THD by a factor of 5 to 7 with respect to a unit with unmatched pairs?
No, I would not make that comparison.

There are many things to consider when interpreting the results.

The initial measurements were taken with a QA-401 and distortion magnifier. That combination is not able to measure as low as our current devices. So its quite possible that we were just measuring the measurement set-up.

The other things is those early measurements were taken without some of the tweaks we have made since those initial measurements were taken. For example the feedback trace tweak.

So you're amplifier will preform just fine if you don't match anything. But if suggest that you look at my list below and consider matching the transistors shown. How far you go down the list is up to you.

1. Q1, Q2
2. Q3, Q4
3. Q5, Q6
4. Q105, Q106
5. Q107, Q108
6. Output Transistors
 
I think many of the builders did not match the output transistors, and you can save yourself some serious money for other components or projects. I myself bought six NJW3281G and six NJW1302G and they where actually pretty close without matching. I also went for 55v rails, a modest 600VA transformer and all 64v caps, saved myself some money. And you know what ? It is still the best amp I ever had.