Thanks for your comments Jeremy. My PS voltage will be about +/-72.5V but I have 2sc4883a/sa1859a which are 180V devices. Are these still no good for 4ohm and +/-72.5V PS?
I also have mje15032/33 so I can use these if 2sa/sc4883a were no good.
Actually, this amp is to drive mid-high range speakers which are: mid (PRV 5mr450ndy)= 8ohm, high=8ohm if i used Vifa or 6ohm if I used DL beryllium drivers. I have separate enclosures for my speakers. LF are neodymium 10" Beyma 2x8ohm drivers in parallel (so 4ohm) working from 100Hz to 400-500z but driven from a different amp, below 100Hz is my active sub with servo.
cheers,
I double checked. You can use 2SC4883A/2SA1859A as drivers at 71V rails and 4 ohms. You can only do this if your outputs 2SC6145/2SA2223 are Y grade, anything else you need the mje15032/33. The 2SC4883A/2SA1859A have better SOA at 71V compared to 2SC4793/2SA1837. Don't even think about using the 2SC4793/2SA1837 in this scenario.
I checked SOA with the heatsink at 60°C
Jeremy
Since you mention "cash grab" I believe this is aimed at me. It would be easier if you referred to people by their name rather than as trolls who are pathetic.
Every member who spent money on this project did it because they wanted to. It's a choice to spend time and money on a hobby or an open project.
I make lots of boards from JLCPCB and never spend US$200 for shipping using 1 week FedEx shipping. I live in South Africa, China is far away. I pay $45 shipping for 60 boards. Shipping from USA to other countries is expensive. Should have shipped straight from the board house for 5 boards, this would have worked out cheaper in my opinion. Enig platting and 2ounce copper isn't needed for prototypes, this saves cost also.
I'm not sure who is emailing you, so no comment on that.
The name of this thread indicated that suggestions are asked for. I suggested something you did not like, Now I'm pathetic troll.
It's amazing how defensive people get when it comes to money.
btw are you stating that the group buys are used to remunerate those that have spent money on this project?
Regards
@Vostro The fees mentioned by @Harry3 were the total shipping cost to send some boards into Greece the shipping fees were actually $220 USD and were mostly import duties. The fees charged by DHL were only $35 USD
You assume that we don't have the intelligence to work out the most economical shipping method.
The boards were actually sent directly from the PCB house and were of course not enig plated and yes we did need 2oz copper as we measured all the trace capacitance and inductance of output stage to include in our ltspice simulation. That information was used to fine tune component values to help us improve the performance and stability margins of the amplifier.
The group buy was held because people wanted to build this amplifier and it was the most economical way we could purchase boards in bulk.
Please don't assume what we did and make judgement on that.
It's actually more polite to actually ask a question that people can respond to.
You assume that we don't have the intelligence to work out the most economical shipping method.
The boards were actually sent directly from the PCB house and were of course not enig plated and yes we did need 2oz copper as we measured all the trace capacitance and inductance of output stage to include in our ltspice simulation. That information was used to fine tune component values to help us improve the performance and stability margins of the amplifier.
The group buy was held because people wanted to build this amplifier and it was the most economical way we could purchase boards in bulk.
Please don't assume what we did and make judgement on that.
It's actually more polite to actually ask a question that people can respond to.
Last edited:
Who is questioning your intelligence? Why would you even write such a sentence. I once ordered using DHL instead of FedEx, in my case it cost me twice the shipping cost, does this mean in not intelligent? not at all, we are all human. Not sure about your motives here.
Since you mention inductance. Could anyone share by how much a pcb trace inductance is affected by copper weight (1oz vs 2oz), 1% ?
It must be nice to have access to measuring equipment that could measure the difference in inductance between 1oz vs 2oz copper. Which frequencies are you using for you impedance measurements?
Regards
I am an outsider here, so take my comment with all its worth, but these last couple of pages do nothing to contribute to this amplifier and, in my opinion, the thread would be better if the posters in question just stopped with the non-contributing nonsense in this thread.
We are here to help each other build a great amplifier and not waste time and energy fighting off abuse. Part of that help has been to attempt preservation of limited parts supply for every person committed enough to invest in the first production run of boards. So sorry if you missed out on the first but be patient, it looks like there might be a second run.
Personally I fear that, collectively, our next design challenge will be to determine the combination of alternate available parts, technologies and rail voltages to get anywhere near to the performance achieved in the first run.
Personally I fear that, collectively, our next design challenge will be to determine the combination of alternate available parts, technologies and rail voltages to get anywhere near to the performance achieved in the first run.
Last edited:
The Sankens have thier hfe grade stamped on the front, R, O, or Y. You want Y grade for the outputs in order to use the 2sc4883a/2sa1859a as drivers at 71V rails. Additionally you can measure the hfe of the outputs and match them up to their compliment. I would screen the drivers for at least a hfe of 100, higher if you can.
Sanken makes a more powerful driver too, 2sc4382/2sa1668 but they are much slower with a Ft of only 20MHz. The mje15032/33 is faster than than these.
Jeremy
Hi guys!
Do the transistors for Q1/2 and Q3/4 have to be "complementary" to each other?
For instance is it bad to use BC559 and BC546 or should it rather be 559/549 or 556/546?
The reason for me asking is that I could not source both recommended components and therefore had to fallback to the alternatives in the bom.
Do the transistors for Q1/2 and Q3/4 have to be "complementary" to each other?
For instance is it bad to use BC559 and BC546 or should it rather be 559/549 or 556/546?
The reason for me asking is that I could not source both recommended components and therefore had to fallback to the alternatives in the bom.
You are correct, there is nothing I could do to contribute. How to contribute to something you can't see?
But since I'm alone in my opinion. Any diyAudio don't care about this amplifier that was promised to them. Why should I.
This is the internet after all and those that are willing to be aggressive and rude seem to rule.
This will be my last post on this thread @DaveFred, I understand that making this circuit diagram available serves in no purpose in contributing to this thread anymore.
Folks:
I just had a very weird experience with Mouser. I ordered most of the parts needed to build my Wolverine amplifier from Mouser in March and they promptly shipped all of the available parts; not surprisingly, a few items were back-ordered. Included in the back-ordered parts were twelve NJW0281G NPN output transistors, which Mouser wrote would ship in a few months. I recently received an update from Mouser indicating that the NJW0281G transistors were going to be further delayed - the revised shipping date was late 2023. Not wanting to wait 17 months to finish this project, I bought replacement output transistors from Newark (both NPN and PNP) and called Mouser to cancel the NJW0281G transistors portion of my outstanding order and return the twelve no-longer-needed NJW0302G PNP transistors they had already sent me.
Mouser's response was that it wasn't worth it for them to issue a shipping label for the NJW0302G transistors (for which I paid over $32 USD); they're issuing a refund for them and don't want them back. I offered to pay the return shipping myself and their response was "don't bother."
Yes, Mouser is one of the more expensive suppliers out there. You certainly pay more for the assurance of genuine parts. But I had forgotten that Mouser's customer service also stands far above that of many of their competitors.
I hope you get your parts. If not, I hope your experience is similar to mine.
Regards,
Scott
I just had a very weird experience with Mouser. I ordered most of the parts needed to build my Wolverine amplifier from Mouser in March and they promptly shipped all of the available parts; not surprisingly, a few items were back-ordered. Included in the back-ordered parts were twelve NJW0281G NPN output transistors, which Mouser wrote would ship in a few months. I recently received an update from Mouser indicating that the NJW0281G transistors were going to be further delayed - the revised shipping date was late 2023. Not wanting to wait 17 months to finish this project, I bought replacement output transistors from Newark (both NPN and PNP) and called Mouser to cancel the NJW0281G transistors portion of my outstanding order and return the twelve no-longer-needed NJW0302G PNP transistors they had already sent me.
Mouser's response was that it wasn't worth it for them to issue a shipping label for the NJW0302G transistors (for which I paid over $32 USD); they're issuing a refund for them and don't want them back. I offered to pay the return shipping myself and their response was "don't bother."
Yes, Mouser is one of the more expensive suppliers out there. You certainly pay more for the assurance of genuine parts. But I had forgotten that Mouser's customer service also stands far above that of many of their competitors.
I hope you get your parts. If not, I hope your experience is similar to mine.
Regards,
Scott
After working for some time with @stuartmp we've finally come up with the final layout for the 5U 400 and the 3U 400 heatsinks which includes both the holes for the UMS template and the Wolverine board! These can already be purchased by those who may need them 
@reddish75 @SRMcGee
@reddish75 @SRMcGeeAttachments
Last edited:
It appears that all my Sankens are Y. NPN/PNP power transistors have hfe very close: 110-120 but hfes of NPN-PNP Sanken drivers are quite different, about 130-140 (NPN) and over 200-210 (PNP). My mje15032/33 have notably lower hfe than Sankens but NPN and PNP are closer. It would be nice to have hfe of NPN-PNP drivers close but not much luck here with Sankens. My Toshiba's 2sa1930.sc5171 on the other hand are quite close with hfe well above 120 but not recommended here.. So what is more important here, close hfe matching of NPN-PNP or speed and other parameters?