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

Member
Joined 2008
Paid Member
Similar questions came to my mind and I kindly ask the team for an advice:
  1. In the sheet 2 of the BOM: Why are there different voltages for linear supplies and SMPS within one line? Is it because the voltages that were actually tested were slightly different for the different supplies?
  2. For my concrete case: I have an 800W/60V SMPS with NJW0281/0302 outputs and MJE1503X drivers on an EF3-3. Would you consider this a reasonable configuration?
Thank you for your help!
SMPS tend to be regulated, and don't sag under load like a linear supply, that's why there are different columns for each. These are all calculated off of the SOA of each of the transistors, this is not a maybe it will be ok guide, these are definite rules to follow, unless you like replacing blown parts down the road. Me personally would not run those outputs with your supply, There are many to chose from that would be safe. Drivers are good though
 
  • Like
Reactions: 3 users
SMPS tend to be regulated, and don't sag under load like a linear supply, that's why there are different columns for each. These are all calculated off of the SOA of each of the transistors, this is not a maybe it will be ok guide, these are definite rules to follow, unless you like replacing blown parts down the road. Me personally would not run those outputs with your supply, There are many to chose from that would be safe. Drivers are good though
Thank you for this very fast and helpful reply (y) (I am not sure but I'd guess that the SMPS values have not been part of the table at the time I ordered the parts - just to my excuse ;))
 
  • Like
Reactions: 1 user
When I built my Wolverine I used an Hypex 1,2kW SMPS at +-60VDC and NJW1302/3281 as output devices, which have better SOA specs, I believe. By the way I'm listening to the amp quite well, without any issue, it becomes barely warm during use and it's truly powerful.

Gaetano.
 
Last edited:
  • Like
Reactions: 1 user
Pre Drivers are ok, MJE Drivers are OK the 3503/1381 are not used for drivers. Outputs I would say NO, I have used MJL4281/MJL4302 on 71v EF3-3 and no issues.


Thank you very much for the prompt response.
Do you think this voltage is too high at +/- 70 volts and I should go to +/-64 volts .These transistors 2sc5359 and 2sa1987 are from amplifier Titan2000 Elektor that was playing there at +/- 70 volts is about the same with 2SC5200 and 2SA1943 Of course there were 4 pairs of output transistors and not three.
 

Attachments

  • ΤΙΤΑΝ2000.jpg
    ΤΙΤΑΝ2000.jpg
    243.4 KB · Views: 265
  • HPIM1894(1).JPG
    HPIM1894(1).JPG
    413.3 KB · Views: 260
  • TITAN 2000(1).jpg
    TITAN 2000(1).jpg
    305.2 KB · Views: 233
  • HPIM1898-1.JPG
    HPIM1898-1.JPG
    494.1 KB · Views: 224
  • IMG_20211224_164602_739.jpg
    IMG_20211224_164602_739.jpg
    382.2 KB · Views: 256
Last edited:
  • Like
Reactions: 1 user
Member
Joined 2008
Paid Member
I saw this on 2 sheet about 2sc5200 was tested at 71V.
Thats onsemi, you said Toshiba? And different ones then you asked about? The Onsemi has a wider SOA then the Toshiba in the same part #, thats way one is rated 71v and the other 64v see attached datasheets and look at the SOA chart you will see. I get confused sometimes when people want to use transistors that are either A not recommended or B not safe for the amount or rail voltage. Sheet 2 is not really a recommendation that you can say oh that will be close enough, its ok I have transistor X and and Y pairs, what is the max voltage period, thats it. 3 pair and 71v is not an issue but you need to use 1 of the 5 transistors that is listed.

I don't mean to come off rude, I just want to make the point very clear to everyone, so in the future they are not replacing damaged components when everything was perfectly laid out from the start.

Cheers
 

Attachments

  • toshiba_2SC5200.PDF
    127.5 KB · Views: 94
  • onsemi_2SC5200.PDF
    238.1 KB · Views: 85
  • Like
Reactions: 1 users
Member
Joined 2016
Paid Member
Similar questions came to my mind and I kindly ask the team for an advice:
  1. In the sheet 2 of the BOM: Why are there different voltages for linear supplies and SMPS within one line? Is it because the voltages that were actually tested were slightly different for the different supplies?
  2. For my concrete case: I have an 800W/60V SMPS with NJW0281/0302 outputs and MJE1503X drivers on an EF3-3. Would you consider this a reasonable configuration?
Thank you for your help!
With a linear supply the voltage drops under load. The amount it drops depends on the VA rating of your transformer.

Switch mode power supplies don't suffer from this.

The transistors information took these facts into account.
 
  • Like
Reactions: 4 users
With a linear supply the voltage drops under load. The amount it drops depends on the VA rating of your transformer.

Switch mode power supplies don't suffer from this.

The transistors information took these facts into account.
Thank you for your reply :)
When I sourced the parts, I had a linear supply in mind. And it was certainly my fault to not re-check the table when I decided to use an SMPS.
I am not at all familiar with SOA considerations, and I have a question wrt. your assumptions:
When calculation the max. SMPS voltages, did you assume a 4ohms load, and DC @ full rail voltage as worst case (and divide the resulting current by 3 or 4)?
 
Hi. I have large stock of BC550C and BC560C, it seems they are the same as the BC549C and BC559C, but with higher collector-emitter breakout voltage.
Can I use them instead of the specified BC549C and BC559C?
Thanks
SB
Same situation here . The question is should the pairs Q1,Q2 and Q3,Q4 be both matched and with each other? Or not? that is Q1,Q2,Q3,Q4 matched all?
 
Last edited:
Member
Joined 2008
Paid Member
I just finished some testing on VAS transistors, CCS transistors, and for fun I wanted to see if there were any benefits to using boutique audiophile capacitors at C1 and how nasty a regular old lytic cap ruins the signal.

First up here are my measurements of some VAS transistor Q13 options, I also measured the C3503D and it performed on par with the C3503E in this position.
Some of these transistors are obsolete or can only be sourced in large quantities but for the most part any distortion at this level of measurement is not audible and you would not be able to tell a difference either way.

From Best VAS transistor to worst would be;
1 / C2682Y
2 / C3503E
3 / C3503D * Should be First Choice as it is easy to find *
4 / TTC004


Vas_Compare.png


For the CCS Transistors Q7,Q8,Q11 it was basically a wash between A1381E / C3503E and TTC004 / TTA004, so if you have the A1381E/C3503E great use those, if not get the easy to find TTC004/TTA004's. I did not test D Grade of the A1381/C3503 and being that the TTC/TTA are easy to find and have a higher gain I would use those, or for anyone who wants to save your E grades the TTC/TTA is great choice.

Here are the different capacitors I tried and measured for C1.

PXL_20230124_182506157.NIGHT.jpg


And Results of those tests

c1_compare1.png


I would not recommend you spend any $$$ on these audiophile caps, not one of them performed better the the Vishay which is the suggested cap on the BOM.

In fact if you want to save money, the BP 47uF cap actually measures the same as the Visahy, and beats all the other caps as well, which was actually a shock to me

c1_compare2.png


Hopefully you guys find this useful and if you have any questions let me know.

Andy
 
  • Like
  • Thank You
Reactions: 18 users