I also got oscillation in mine. I made my own DIY boards and I first got 700kHz at 70v peak when I used intersil IRFP250's then when I used international rectifier ones, I got 500kHz at 15V peak.
my solution was to connect a 100pF cap across the gate and drain terminals of the Q13 mosfet.
I also added an additional pair of output stages.
my solution was to connect a 100pF cap across the gate and drain terminals of the Q13 mosfet.
I also added an additional pair of output stages.
Commentable Thoughts.
Does any one have tested the ANTHONY'S N-channel AMP FOR CROSS-CONDUCTION between upper and lower mosfets.
If u put the full drive signal at input and the output is open that is no load connected the amp starts self heating due to slow turn off times between mosfets and sometimes self destructs.
Try this test!
Regards,
AmpMan
Does any one have tested the ANTHONY'S N-channel AMP FOR CROSS-CONDUCTION between upper and lower mosfets.
If u put the full drive signal at input and the output is open that is no load connected the amp starts self heating due to slow turn off times between mosfets and sometimes self destructs.
Try this test!
Regards,
AmpMan
i built one n-channel too. the pcb is something like the one posted here. the problem is that it oscillates the hell out of it, it heats up like a toaster in no time. So what can i do about that? I tried a 100nF cap in series with 10ohm at the output but no succes. Should i make the gate connections shorter somehow? I get a sinusoidal oscillation in the 1-10 mhz range and if y touch the output it stops my cdplayer from working. hmm, what i've built here, amp or transmitter? 
. My transistors are irf530. Thanks for help !
. My transistors are irf530. Thanks for help !
KillerOPS said:i built one n-channel too. the pcb is something like the one posted here. the problem is that it oscillates the hell out of it, it heats up like a toaster in no time. So what can i do about that? I tried a 100nF cap in series with 10ohm at the output but no succes. Should i make the gate connections shorter somehow? I get a sinusoidal oscillation in the 1-10 mhz range and if y touch the output it stops my cdplayer from working. hmm, what i've built here, amp or transmitter?
1st: Try to decrease the value of R31, to the half it! This will increase the close loop gain. If the oscillation caused by the poor loop stability it will help.
2nd: Try to increase the gatestoppers at the output devices, if the 1st doesn't help.
sajti
i started building another pcb because the modifications didn't help me. i hope it will solve my problems. does anyone know another n-channel only mosfet amplifier schematic around 200w?
I am also interested in some hints about the pcb, which track to be bigger/smaler/closer/shorter, component placements and so on. i want to build it for a car amp, and i need the output mosfets to be on one side of the board. i would like to use irf540 for it because it's cheap. any comments?
I am also interested in some hints about the pcb, which track to be bigger/smaler/closer/shorter, component placements and so on. i want to build it for a car amp, and i need the output mosfets to be on one side of the board. i would like to use irf540 for it because it's cheap. any comments?
Ok, i decided not to redo that pcd, instead choose another schematic. What do you think about this? http://xedox.de/120_watt.jpg
What modifications does it needs in order to use irf540 as output devices? Only N-channel. I used a BD238 and BD237, with 470 ohm emitter resistors and 470 ohm irf540 gate resistor instead of those bdx-es, but it doesn't sounds too good. THere are a lot of highs. And i changed the 1000uF cap to 47uF and it no longer pushes the cone out of the speaker at low level input. What am i missing? Also, i would like to use 4-5 devices per rail, at around +- 60-70v.
So please tell me, is it a good idea to stick with this one or the original N-Channel? I'm a little afraid of that lm317 and all the things around it in the n-channel amp.
later edit: wrong url in first post
What modifications does it needs in order to use irf540 as output devices? Only N-channel. I used a BD238 and BD237, with 470 ohm emitter resistors and 470 ohm irf540 gate resistor instead of those bdx-es, but it doesn't sounds too good. THere are a lot of highs. And i changed the 1000uF cap to 47uF and it no longer pushes the cone out of the speaker at low level input. What am i missing? Also, i would like to use 4-5 devices per rail, at around +- 60-70v.
So please tell me, is it a good idea to stick with this one or the original N-Channel? I'm a little afraid of that lm317 and all the things around it in the n-channel amp.
later edit: wrong url in first post
Banned
Joined 2002
What is wrong with the original n-channel ? I have one .. soon to be another as there great work horse amps .. i use this amp allot and put it to hell and back : O ) its currently sitting in a box with out the out-puts.. because i blew up it up simple error moving it to a new heat-sink.. : O )
killerops, I can suggest you try my previous solution to get rid of the oscillation. http://www.diyaudio.com/forums/showthread.php?postid=440606#post440606
djQUAN: i can't even remember how many tries i had with caps around the gates of the o/p and all the transistors. It doesn't work.
sajti: tried that, doesn't work
richie00boy: yes, the link is ok. Instead of the output transisors on that schematic i used something like a darlington connection. Up: irf540 and bd237: the collector of the bd and the drain of the irf goes to +40v. the emitter of the bd goes trough a 470 ohm resistor to the source of the mosfet and from the emitter is another 470 ohm resistence that goes to gate. the base of the bd goes to the 270ohm resistence.
sajti: tried that, doesn't work
richie00boy: yes, the link is ok. Instead of the output transisors on that schematic i used something like a darlington connection. Up: irf540 and bd237: the collector of the bd and the drain of the irf goes to +40v. the emitter of the bd goes trough a 470 ohm resistor to the source of the mosfet and from the emitter is another 470 ohm resistence that goes to gate. the base of the bd goes to the 270ohm resistence.
of course is different. it's the same thing, but with bd238 and the connections are not the same.
anyway, it doesn't maater, this works. the amplifier works, but it doesn't sounds good. i suppose some of the parts's values must be tweaked for mosfet output. but i have no idea which ones. if anyone can help...
anyway, it doesn't maater, this works. the amplifier works, but it doesn't sounds good. i suppose some of the parts's values must be tweaked for mosfet output. but i have no idea which ones. if anyone can help...
Questions on PCB and Heatsinking
Which version, as in date, of schematic did you use?
Was this like:
---------+----------> + speaker out
|
|
|
\
/
\
/ 5R6
\
/
|
|
---
--- 220n0
|
|
|
Signal
GND
Any ideal if you used IRFP240, IRFP250 or IRFP250N you would not have experienced the oscillations? I am thinking of building a few of the n-channel amps. The possible output driver considerations will be:
1) 4 pairs of IRFP240's or IRFP244's (108W RMS at >= 2 ohms)
2) 2 pairs of IRFP240's or IRFP244's (54W RMS at >= 2ohms)
3) 1 pair IRFP240's or IRFP244's (25W rms at >= 2 ohms)
4) 1 pair IRF640N's (25W rms at >= 2ohms)
5) 1 pair IRF630's (12W RMS at >= 2 ohms)
I assume due to the lower Ciss (Gate Capacitance) the RC Network you found necessary may not be necessary at the >1400 Ciss for the output devices I am interested in?
I like your PCB board. I have been trying to do a hand layout using actual scale of components in a graphics program. I use Linux which limits my choice of PCB design software. Eagle has a Linux based program, but it is buggy as anything and seems to have some poor design. I have no idea if the Windows based Eagle is as buggy and design challenged as the Linux one is. I tried various Linux CAD based programs, but they all have bugs or serious useability issues. They will catch up in time, just not for my very modest current needs. I found TurboCADD to be best CAD software on Windows when I used Windows for very brief time on my system.
I noticed you have a PCB based heatsink. I am not sure, but seems the device for the PCB heatsink my be a plastic LM317. The heatsink you are using is exacly same one I have bought and have been working into the PCB for both IRF610's and a plastic LM317T/TF. Can you tell me how warm to touch the LM317 gets on your n-channel amp? How warm do each of the IRF610's get to touch? Can you confirm what device you chose to use a PCB heaksink for, as I am assuming for moment it is the plastic LM317?
Any chance I could have copy of the top and bottom image, with board width and lenght in case when I post my hand layout PCB it is met with grave dispair or I find once I build one it has serious problems? I may not get the hand design posted for week or two. I am at the hardest part, working out the final last few components always seems to take the longest. In this case Q11, R2, R6, R26, ZD4 with the ouput drivers and associated resistors. The output drivers and related resistors will not pose that big a challenge as this should be standard layout used on many past PCB layouts. I will then see what areas I can optimize. I know some I can and will be easy I think, a few others may be more of challenge to optimize.
Regards,
John L. Males
Willowdale, Ontario
Canada
29 December 2004 09:26
Pelle said:
Which version, as in date, of schematic did you use?
Was this like:
---------+----------> + speaker out
|
|
|
\
/
\
/ 5R6
\
/
|
|
---
--- 220n0
|
|
|
Signal
GND
Any ideal if you used IRFP240, IRFP250 or IRFP250N you would not have experienced the oscillations? I am thinking of building a few of the n-channel amps. The possible output driver considerations will be:
1) 4 pairs of IRFP240's or IRFP244's (108W RMS at >= 2 ohms)
2) 2 pairs of IRFP240's or IRFP244's (54W RMS at >= 2ohms)
3) 1 pair IRFP240's or IRFP244's (25W rms at >= 2 ohms)
4) 1 pair IRF640N's (25W rms at >= 2ohms)
5) 1 pair IRF630's (12W RMS at >= 2 ohms)
I assume due to the lower Ciss (Gate Capacitance) the RC Network you found necessary may not be necessary at the >1400 Ciss for the output devices I am interested in?
I like your PCB board. I have been trying to do a hand layout using actual scale of components in a graphics program. I use Linux which limits my choice of PCB design software. Eagle has a Linux based program, but it is buggy as anything and seems to have some poor design. I have no idea if the Windows based Eagle is as buggy and design challenged as the Linux one is. I tried various Linux CAD based programs, but they all have bugs or serious useability issues. They will catch up in time, just not for my very modest current needs. I found TurboCADD to be best CAD software on Windows when I used Windows for very brief time on my system.
I noticed you have a PCB based heatsink. I am not sure, but seems the device for the PCB heatsink my be a plastic LM317. The heatsink you are using is exacly same one I have bought and have been working into the PCB for both IRF610's and a plastic LM317T/TF. Can you tell me how warm to touch the LM317 gets on your n-channel amp? How warm do each of the IRF610's get to touch? Can you confirm what device you chose to use a PCB heaksink for, as I am assuming for moment it is the plastic LM317?
Any chance I could have copy of the top and bottom image, with board width and lenght in case when I post my hand layout PCB it is met with grave dispair or I find once I build one it has serious problems? I may not get the hand design posted for week or two. I am at the hardest part, working out the final last few components always seems to take the longest. In this case Q11, R2, R6, R26, ZD4 with the ouput drivers and associated resistors. The output drivers and related resistors will not pose that big a challenge as this should be standard layout used on many past PCB layouts. I will then see what areas I can optimize. I know some I can and will be easy I think, a few others may be more of challenge to optimize.
Regards,
John L. Males
Willowdale, Ontario
Canada
29 December 2004 09:26
LM317 Pinouts
richie00boy,
Thanks for heads up. I been aware of these nunsances based on many a amp notes and postings here of differences in pinouts with alternate transistors. I will just have to try to keep your reminder in mind if I use different part again.
As side note, being very aware of your point as I try to layout my PCB by hand I have had darnest problem with the 2SC3298. I cannot find a datasheet nor not even a hint of what the pinout of the 2SC3298 is. I have searched the internet and ftp high and low in very creative ways for 2 days. Seems one can still get the 2SC3298 from all kinds of sources, just the any details of the pinouts cannot be found nore deduced from any PCB that uses the 2SC3298 that I can find. I am not sure what the redeeming qualities of the 2SC3298 are. LBHajdu had suggested replacing the 2SC3298 with an IRF610 in Post #14:
http://www.diyaudio.com/forums/showthread.php?postid=277285#post277285
I am not sure if using an IRF610 is possible or what the tradeoffs are, if any. While looking for data on the 2SC3249 I did find
http://www.bmm-electronics.com/Product.asp?Product_ID=679
2SC4159
NPN transistor 180V 1.5A 15W 100MHz TO220ML
Complementary to 2SA1606
Replacement for 2SC3298 and 2SC4370
I am not sure if the 2SC4159 is at least as good as the 2SC3298, but in the absence of any 2SC3298 data and pinout information I have designed the PCB layout to use a 2SC4159 for now hoping any differences in pinout use to either 2SC3298 or IRF610 do not exist.
Regards,
John L. Males
Willowdale, Ontario
Canada
30 December 2004 00:45
richie00boy,
Thanks for heads up. I been aware of these nunsances based on many a amp notes and postings here of differences in pinouts with alternate transistors. I will just have to try to keep your reminder in mind if I use different part again.
As side note, being very aware of your point as I try to layout my PCB by hand I have had darnest problem with the 2SC3298. I cannot find a datasheet nor not even a hint of what the pinout of the 2SC3298 is. I have searched the internet and ftp high and low in very creative ways for 2 days. Seems one can still get the 2SC3298 from all kinds of sources, just the any details of the pinouts cannot be found nore deduced from any PCB that uses the 2SC3298 that I can find. I am not sure what the redeeming qualities of the 2SC3298 are. LBHajdu had suggested replacing the 2SC3298 with an IRF610 in Post #14:
http://www.diyaudio.com/forums/showthread.php?postid=277285#post277285
I am not sure if using an IRF610 is possible or what the tradeoffs are, if any. While looking for data on the 2SC3249 I did find
http://www.bmm-electronics.com/Product.asp?Product_ID=679
2SC4159
NPN transistor 180V 1.5A 15W 100MHz TO220ML
Complementary to 2SA1606
Replacement for 2SC3298 and 2SC4370
I am not sure if the 2SC4159 is at least as good as the 2SC3298, but in the absence of any 2SC3298 data and pinout information I have designed the PCB layout to use a 2SC4159 for now hoping any differences in pinout use to either 2SC3298 or IRF610 do not exist.
Regards,
John L. Males
Willowdale, Ontario
Canada
30 December 2004 00:45
n-channel questions
Finaly i decided to build A. Holtons n-channel. When I visited his page i realized that the pcb is no longer avalible. There is a new updated version comming in the future. So, i decided to make my own pcb and did so with the silkscreen that is at the beginning of this page.
I can not get it to work. I get negative rail power at the output terminal. I am using one pair of IRFP260. I have crosschecked the schematic against the pcb many times and I can not find any faults. I have changed components (all transistor and diods) and I am using the LM317T (it says 317K in his manual). I test the amp on +- 35v and get -33v comparing to gnd
You that have made this amplifier, is there any error in the schematic ?
Anything that is important that not mentioned in the manual
Anyone that have any sugestions?
I would be greatful if any of you experts could help me
Regards,
Bengt
Finaly i decided to build A. Holtons n-channel. When I visited his page i realized that the pcb is no longer avalible. There is a new updated version comming in the future. So, i decided to make my own pcb and did so with the silkscreen that is at the beginning of this page.
I can not get it to work. I get negative rail power at the output terminal. I am using one pair of IRFP260. I have crosschecked the schematic against the pcb many times and I can not find any faults. I have changed components (all transistor and diods) and I am using the LM317T (it says 317K in his manual). I test the amp on +- 35v and get -33v comparing to gnd
You that have made this amplifier, is there any error in the schematic ?
Anything that is important that not mentioned in the manual
Anyone that have any sugestions?
I would be greatful if any of you experts could help me
Regards,
Bengt
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.