Hi everyone !
So, I REALY want to build this amp:
http://www.astro.uu.se/~marcus/private/m250.html
But, there's no PCB! Can you help me to design this PCB ?
What program should I use ? Anyone here have this PCB done?
Another thing ! The output stage transistors, here are very expensive... The transistors used in this project are MJ15024, that is 16A 200V.
Why 200V if the amp works with 55+55v ?
So, can I replace them with TIP3055 or TIP35? The first is 15A and the second 25A, they are 100v.
If I replace with 3055, I'll add more one pair, because it's 15A, and the original is 16A...
The output stage is a half PNP and a half NPN, and if the supply is 55 + 55v, each stage works with 55v right? So, there's no problem using 100v transistors right?
Thaaaaankksssss !!!!!!!
So, I REALY want to build this amp:
http://www.astro.uu.se/~marcus/private/m250.html
But, there's no PCB!
Can you help me to design this PCB ? What program should I use ? Anyone here have this PCB done?
Another thing ! The output stage transistors, here are very expensive... The transistors used in this project are MJ15024, that is 16A 200V.
Why 200V if the amp works with 55+55v ?
So, can I replace them with TIP3055 or TIP35? The first is 15A and the second 25A, they are 100v.
If I replace with 3055, I'll add more one pair, because it's 15A, and the original is 16A...
The output stage is a half PNP and a half NPN, and if the supply is 55 + 55v, each stage works with 55v right? So, there's no problem using 100v transistors right?
Thaaaaankksssss !!!!!!!
-_nando-_ said:
Another thing ! The output stage transistors, here are very expensive... The transistors used in this project are MJ15024, that is 16A 200V.
Why 200V if the amp works with 55+55v ?
So, can I replace them with TIP3055 or TIP35? The first is 15A and the second 25A, they are 100v.
If I replace with 3055, I'll add more one pair, because it's 15A, and the original is 16A...
The output stage is a half PNP and a half NPN, and if the supply is 55 + 55v, each stage works with 55v right? So, there's no problem using 100v transistors right?
Thaaaaankksssss !!!!!!!
The output devices have to operate from rail to rail, 55 + 55 = 110V!. So you may be pushing it here. You also need to have similar Hfe and be mindful of safe operating area. SOA is the parameter that determines how many devices you need, not current rating. TIP35 is a rugged device, but I think the Hfe is lower.
TIP35C has a Vce of 100v max so this too could expire. You can use almost any transistors for the output provided that;
they have a Vce of at least 120v
they have a gain of at least 20 or so
you can get them in P and N
they have a combined power rating of at least 800 w per rail
are the right package (or one that you can use)
are not a darlington (for simplicity)
But you know what, Ive had a look at the cct and there are better amps out there, so if you don't have a board yet and haven't started, why don't you look for alternatives, plenty around here you can use.
Cheers
they have a Vce of at least 120v
they have a gain of at least 20 or so
you can get them in P and N
they have a combined power rating of at least 800 w per rail
are the right package (or one that you can use)
are not a darlington (for simplicity)
But you know what, Ive had a look at the cct and there are better amps out there, so if you don't have a board yet and haven't started, why don't you look for alternatives, plenty around here you can use.
Cheers
Hi,
what current can the Tip take when dropping 55V across CE?
That is the output device current limit when driving a speaker (reactive) load.
Using an 8ohm speaker with a 6r minimum resistance as an example.
The peak current from your amp will be (55-5)/6 =8.3Apk.
The derating factor for your Tip will depend on the transistor casing temperature. This in turn depends on the heatsink size and airflow and how much power the amp has recently been pumping out (both quiescent+output).
If the Tc = 50degC then derating is likely to be 0.8
Tip current = SOAR current at 55V * 0.8.
number of devices = output current/ tip current = 8.3/SOAR/0.8 = a lot of devices.
If you use a higher voltage rating device then usually the SOAR goes up dramatically and you can then reduce the number of parallel devices. MJL15024 has a max 4.5A at 55V. derated = 3.6A. Number of devices = 8.3/3.6 = 2.3 therefore use 3pairs for an 8 ohms load with any phase angle and 5 pairs for 4 ohm load at any phase angle. With accurate V & I analysis you may be able to trim off 1 pair from each of the example calculations resulting in a robust 2pair 8ohm amp or 4 pair 4ohm amplifier.
Using Tips is probably going to be a non starter. Try working out the numbers. Unfortunately design involves maths, change the design and you change the arithmetical results. Avoiding the maths may result in a blown amp.
what current can the Tip take when dropping 55V across CE?
That is the output device current limit when driving a speaker (reactive) load.
Using an 8ohm speaker with a 6r minimum resistance as an example.
The peak current from your amp will be (55-5)/6 =8.3Apk.
The derating factor for your Tip will depend on the transistor casing temperature. This in turn depends on the heatsink size and airflow and how much power the amp has recently been pumping out (both quiescent+output).
If the Tc = 50degC then derating is likely to be 0.8
Tip current = SOAR current at 55V * 0.8.
number of devices = output current/ tip current = 8.3/SOAR/0.8 = a lot of devices.
If you use a higher voltage rating device then usually the SOAR goes up dramatically and you can then reduce the number of parallel devices. MJL15024 has a max 4.5A at 55V. derated = 3.6A. Number of devices = 8.3/3.6 = 2.3 therefore use 3pairs for an 8 ohms load with any phase angle and 5 pairs for 4 ohm load at any phase angle. With accurate V & I analysis you may be able to trim off 1 pair from each of the example calculations resulting in a robust 2pair 8ohm amp or 4 pair 4ohm amplifier.
Using Tips is probably going to be a non starter. Try working out the numbers. Unfortunately design involves maths, change the design and you change the arithmetical results. Avoiding the maths may result in a blown amp.
Hi Quasi,
don't give in just on a whim.
100Vce0 and 120Vce0 could just about do the job but he would need an awfull lot of them. Let him do the arithmetic and he'll learn from the experience. Lots of cheap components or fewer more expensive components are, in this case, equivalent.
A better solution to allow the low voltage transistors to work is use lower voltage PSU rails, try +-40Vdc or +-45Vdc. Then repeat the arithmetic.
don't give in just on a whim.
100Vce0 and 120Vce0 could just about do the job but he would need an awfull lot of them. Let him do the arithmetic and he'll learn from the experience. Lots of cheap components or fewer more expensive components are, in this case, equivalent.
A better solution to allow the low voltage transistors to work is use lower voltage PSU rails, try +-40Vdc or +-45Vdc. Then repeat the arithmetic.
MJ15015/16 are 120V, 15A, 180W with a full 3A SOA at 60V, and about 1/2 the cost of MJ15024/25.
Crown and Carver used these in several models. Crown got 240W with only two pair, but I would recommend more at that power level. A different model Crown with four pair of these did 335W 20~20Khz, 400W 1Khz.
An amp I did with three pair of these on ±57V has been running at over 200W into 4 ohms for over 20 years. This amp was fan cooled for small size. I used the Pass A40 as the driver board, and wired the drivers and outputs point-to-point on their sockets.
The A40 driver board is still available for only $6 from AudioXpress. The outputs should be available for less than $2 each.
I give alternate transistor numbers for this project, and a link to the board source, at the Parts Express tech forum. Search on A40 and djk.
Crown and Carver used these in several models. Crown got 240W with only two pair, but I would recommend more at that power level. A different model Crown with four pair of these did 335W 20~20Khz, 400W 1Khz.
An amp I did with three pair of these on ±57V has been running at over 200W into 4 ohms for over 20 years. This amp was fan cooled for small size. I used the Pass A40 as the driver board, and wired the drivers and outputs point-to-point on their sockets.
The A40 driver board is still available for only $6 from AudioXpress. The outputs should be available for less than $2 each.
I give alternate transistor numbers for this project, and a link to the board source, at the Parts Express tech forum. Search on A40 and djk.
But as the problem related by them, if I use a lower voltage transistor (at least 120v), I'll need more pairs...
ML15004/3 are 20A 140v, I think it work well in this amp, but as I don't know the maths, I'll not try it without any math...
I found the original transistors with a "low" price, but I found 2SC 3281 for the HALF of 15024...
2SC 3281 is a 200v 15A 150W transistor, thermal resistence 0,83 °C/W, Hfe:
lc = 8.0A Vce = 5.0v min 35 max 160
MJ15004 is a 250v 16A 250W transistor, thermal resistence 0,70 °C/W, hfe:
lc = 8.0A Vce = 4.0v min 15 max 60
Wow, not easy...
Can I replace with 3281 ? Do I need to add more pairs because the lower current and watts ? How many?
I've found this amp (c200):
http://www.ampslab.com/c200.htm
Is it good? M250 is better right? And about sound quality, M250 is very better right?
As you see, I want to build M250...
ML15004/3 are 20A 140v, I think it work well in this amp, but as I don't know the maths, I'll not try it without any math...
I found the original transistors with a "low" price, but I found 2SC 3281 for the HALF of 15024...
2SC 3281 is a 200v 15A 150W transistor, thermal resistence 0,83 °C/W, Hfe:
lc = 8.0A Vce = 5.0v min 35 max 160
MJ15004 is a 250v 16A 250W transistor, thermal resistence 0,70 °C/W, hfe:
lc = 8.0A Vce = 4.0v min 15 max 60
Wow, not easy...
Can I replace with 3281 ? Do I need to add more pairs because the lower current and watts ? How many?
I've found this amp (c200):
http://www.ampslab.com/c200.htm
Is it good? M250 is better right? And about sound quality, M250 is very better right?
As you see, I want to build M250...
If they are lower price, they will undoubtedly be fake, especially as Toshiba haven't made them since 2000.
The M250 has an over exaggerated output stage, as Marcus says, he wanted an amp he could run a party with at loud volumes and not have it die. You could probably get 250W out of 3 pairs just fine.
The M250 has an over exaggerated output stage, as Marcus says, he wanted an amp he could run a party with at loud volumes and not have it die. You could probably get 250W out of 3 pairs just fine.
Only toshiba made them? 
Look: http://www.alldatasheet.com/view.jsp?Searchword=2SC
There's MOSPEC and WINGSHING manufactures, and probabily they're still making 3281...
The most funny is, toshiba didn't appear at datasheet catalog and alldatasheet as manufacturer of 3281
HOW
So, I'll use 4 pairs
Changing topic...
Is the sound of M250 better than c200 or c300 ?
Thanks !!!
Look: http://www.alldatasheet.com/view.jsp?Searchword=2SC
There's MOSPEC and WINGSHING manufactures, and probabily they're still making 3281...
The most funny is, toshiba didn't appear at datasheet catalog and alldatasheet as manufacturer of 3281
HOW So, I'll use 4 pairs
Changing topic...
Is the sound of M250 better than c200 or c300 ?
Thanks !!!
Toshiba made the ones that were real and would work to full spec. Wingshing, well that sounds Chinese and pretty much all of the Chinese clones of this transistor (especially when a lot cheaper than the real 3281) are known to just go *bang* when pushed hard enough. No comment on Mospec.
Unless you can guarantee they aren't fakes, you will be wasting your money as they will just blow on switch-on, or when you put some load on the amp... and you will get collateral damage often blowing all the other transistors in teh circuit. It's just not worth it.
Get some good quality transistors, the MJ's aren't that expensive really. Make sure they are OnSemi brand, if they are Motorola brand they will probably be fakes in your country. Motorola no longer make transistors and spun that division off - as On Semiconductor.
Unless you can guarantee they aren't fakes, you will be wasting your money as they will just blow on switch-on, or when you put some load on the amp... and you will get collateral damage often blowing all the other transistors in teh circuit. It's just not worth it.
Get some good quality transistors, the MJ's aren't that expensive really. Make sure they are OnSemi brand, if they are Motorola brand they will probably be fakes in your country. Motorola no longer make transistors and spun that division off - as On Semiconductor.
Heh, realy not good this situation...
How can I test the transistor to see if they're genuine or not?
I realy don't want make a power amp with fake transistors...
And, do you know answer me, if that c200 and c300 from AMPSLAB are good ? Is M250 a real good amp, I mean, witch one is better to spend MY MONEY ?
Thank you a lot !
How can I test the transistor to see if they're genuine or not?
I realy don't want make a power amp with fake transistors...
And, do you know answer me, if that c200 and c300 from AMPSLAB are good ? Is M250 a real good amp, I mean, witch one is better to spend MY MONEY ?
Thank you a lot !
That is the million dollar question. Generally, you have to buy from a reputable supplier, and if they are much cheaper than other similar/same transistors from other suppliers, they will be fakes.
Don't use unknown/chinese brands. Stick with Toshiba, OnSemi, etc..
Originally posted by -_nando-_
And, do you know answer me, if that c200 and c300 from AMPSLAB are good ? Is M250 a real good amp, I mean, witch one is better to spend MY MONEY ?
Never heard either of them but they are pretty similar in topology. I would say the M250 is better as the MJE340/350 are more suitable for VAS than the TIP29/30 used by the C200, but that's only one piece of the puzzle. Personally i prefer fully complementary amps but thats more complex and expensive.
I have run the MJ15024/25 with +/- 100v rails. 4 pairs were enough and I got a lot of power out of it at 8 ohms. Never blew the 24/25's because of SOAR limitations but did becuase of my sinks being non-thermal compensated and having a high bias, caused thermal runaway.
The PSU was not overbuilt so at full load, I'd say the rails would sag to 88vdc.
Heck even ran it bridged with 85v rails and got close to 1000watts into 4 ohms, no problems. They are very rugged devices.
Disclaimer: Just because I got away with it does not mean I suggest doing it. Just that for high power applications, the 24/25's were pretty good in their day.
The PSU was not overbuilt so at full load, I'd say the rails would sag to 88vdc.
Heck even ran it bridged with 85v rails and got close to 1000watts into 4 ohms, no problems. They are very rugged devices.
Disclaimer: Just because I got away with it does not mean I suggest doing it. Just that for high power applications, the 24/25's were pretty good in their day.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.