DIYA store "Wolverine" (Son of Badger) .... suggestions ??

Member
Joined 2016
Paid Member
Yes, with a 3 pair output stage not just any transistor will work on 70V rails. These plots show the temperature derated SOA(per transistor) with a heatsink temperature of 60°C per driving 4 ohms at 45° phase angle (which means the real part of the load is 2.83 ohms and the reactive part is 2.83 ohms) using 3 pair. These plots also take into account the rails sagging by 10% which would be the equivalent of a 1200VA transformer for two channels. This is why we made a 4 pair option, so more than a select few transistors can run on 70V rails.

Jeremy
Thanks @jjs once again @AudioSan has assumed that we haven't done any research or calculations on this.
@AudioSan please know we have looked at the this from dare I say it, every possible angle. Once again I ask you to ask us a question before trying to find a issues with the work the team has done. Through asking questions people are more inclined to work with you and not become defensive. Unless of course you don't need to ask questions because you know everything. If that's the case why are you subscribed to this thread? What are you contributing?
I thought your last post was a few days ago?
 
  • Like
Reactions: 3 users
It's a lot more to it then cooling. SOA, Outputstage capacity, railvoltage and load. Clipping becomes an isue when driven like your calculations.
you can try driving a 3 pair 150W BJT output at +/-70V with 4ohm load (that dips into 3ohm) as hard as hard gets continuously for 8 hours. I bet it will not survive even if you cool it with liquid nitrogen.
Mine is air cooled, uses 70V rails, drives 4ohm loads and has run for well over 8 hours continuously. It gets hot, but that's expected at 150Wrms... Hasn't stopped working yet
 
  • Like
Reactions: 1 users
Brettn56. You drive it at the edge of clipping for 8 hours straight using 3 pair of 150W output BJT's? if not, your post is useless. As this is the senerio discussed.
Of cource any amp at any voltage (within the parts limit) will survive having a 4ohm load connected. As long as the volume level is within reason.
 
Stuartmp. my last post few days ago? please find that post for me. Maybe you learn that it's either the wrong tread and week/month. Or the wrong person.
On a side note. This is about general rule of tumbs and senerios to take into account. So i ask you. read once more.
But hey, since you don't want anyone that is not on the groupbuy to post anything. After all, no one else can contribute to any spesific here
 
Member
Joined 2016
Paid Member
Stuartmp. my last post few days ago? please find that post for me. Maybe you learn that it's either the wrong tread and week/month. Or the wrong person.
On a side note. This is about general rule of tumbs and senerios to take into account. So i ask you. read once more.
But hey, since you don't want anyone that is not on the groupbuy to post anything. After all, no one else can contribute to any spesific here
Apologies @AudioSan I had you confused with Vostro.
 
Brettn56. You drive it at the edge of clipping for 8 hours straight using 3 pair of 150W output BJT's? if not, your post is useless. As this is the senerio discussed.
Of cource any amp at any voltage (within the parts limit) will survive having a 4ohm load connected. As long as the volume level is within reason
I'm not on this forum to argue, taunt, mislead or be a general nuisance. I'm here to offer advice when necessary, and I'm here to learn.

I'm not going to argue with you here as I can already clearly see where this is going.

I work in electronics installation/design, and have an electronic engineering degree, computer systems degree and many others. I've never used or seen liquid nitrogen used for cooling in consumer grade amplifier or linear regulated power supplies.

I think your comment about requiring liquid nitrogen is the true definition of a useless post.

I you don't want to believe that I and others have built and tested the amp in question that's fine.

Please don't go misleading others on the forum. Please share your results from the Amp that you have built, run it at 76V rails, put a 3ohm resistive/inductive load on it and run it for 12hours at 150WRMS. Post your results here and either prove or disprove all parties involved.
If you aren't willing to do that, then go annoy another forum elsewhere.
 
  • Like
Reactions: 2 users
+/-76Vdc rails loaded at 3 ohm gives you around 650W average and 1300W peak. With 3 pair of 150W BJT's??? And you gonna push those figures for 8hours straight. Right. A good role of thumb is for power output to stay within 1/5 of the total capacity of the outputstage. 150x6=900/5=180W average. But hey. you could try 650W average and see if the 900W outputstage can handle it.
 
Member
Joined 2016
Paid Member
Incase your wondering D and E are different HFE grades with E being higher. Ideally the pnp and npn pairs should have the same hfe grade. Check the datasheet for details. And just so you know these were not sourced from china. The are original toshiba 2SA1381E and 2SC3503E. They came with certification and were purchased from a reputable company in the USA
 

Attachments

  • Screenshot_20220519-224717_Photos.jpg
    Screenshot_20220519-224717_Photos.jpg
    197.2 KB · Views: 121
  • Screenshot_20220519-224653_Photos.jpg
    Screenshot_20220519-224653_Photos.jpg
    234 KB · Views: 116
Thanks Jeremy,

So the plan is :

Outputs: 2SC5359/2SA1987
Drivers: MJE15032G/MJE15033G
Pre-Drivers: KSC3503DS/KSA1381ESTU
+/- 70V rails
Heatsinks: 400 x 150 x 50 mm

Is it OK ?

Let me double check and Ill get back to you. I don't have those output transistors loaded in my SOA spreadsheet. This takes a a little bit of time to read the log graphs. How many output pairs?

Jeremy
 
  • Like
Reactions: 1 user
I ran some dissapation numbers for 4 ohms driving a reactive load with 10dB peak to average music. Here are the max rail voltages using standard rectified and filtered power supply with about 10% sag under load. You can do more on the smaller chassis but you'll have to put a fan on the sinks.

Here is a good rule of thumb

3U 40x120x400mm ~50V rails
4U 40x165x400mm ~60V rails
5U 40x210x400mm ~70V rails

Jeremy

Sorry Guys I cut a pasted the rail data for the 300mm to the 400mm deep heatsinks. Here are the corrected values.

3U 0.40°C/W, 40×120×300mm ~50V rails
4U 0.31°C/W, 40x165x300mm ~60V rails
5U 0.18°C/W, 40x210x300mm ~70V rails

3U 0.25°C/W, 40x120x400mm ~60V rails
4U 0.19°C/W, 40x165x400mm ~70V rails
5U 0.14°C/W, 40x210x400mm ~80V rails

Jeremy
 
  • Like
Reactions: 2 users
Thanks Jeremy,

So the plan is :

Outputs: 2SC5359/2SA1987
Drivers: MJE15032G/MJE15033G
Pre-Drivers: KSC3503DS/KSA1381ESTU
+/- 70V rails
Heatsinks: 400 x 150 x 50 mm

Is it OK ?
Jeremy, it is 0,3 K/W

0.3°C/W might be a little small for 70V rails but falls in around my 4U 400mm deep example (size wise). It will probably be ok if you're not running for long periods. If you do need to run it for long periods at full output a fan on the sinks will be necessary and will be so loud you won't hear the fan. Really this applies to everyone.

The 2SC5359/2SA1987 are basically the best 2SC5200/2SA1943. The SOA lines are almost on top of each other for 4 pair. Theses will run roughly 5°C cooler because of the 180W dissaption limit versus the 150W limit of the 2SC5200/2SA1943. Your drivers are not an issue, but I'd check/screen their hfe though.

Jeremy
 
Last edited:
  • Like
Reactions: 1 user