Hi John,
Fast fuses are my choice.
PSU: ~ +/-70V,400VA transformer.
Speaker load: not important, I short the amp with bad speaker cable. I repeted this on both chanels, and both has blown only fuses. In few occasions before I played the amp in full power in a few hours, and I haven't got any problems with fuses.
It's extraordinary to me that amp survived. When I found that I shorted the amp, I was starting to say good bye to mosfets.
I guess I was wrong. Good for me😉
Regards...
Fast fuses are my choice.
PSU: ~ +/-70V,400VA transformer.
Speaker load: not important, I short the amp with bad speaker cable. I repeted this on both chanels, and both has blown only fuses. In few occasions before I played the amp in full power in a few hours, and I haven't got any problems with fuses.
It's extraordinary to me that amp survived. When I found that I shorted the amp, I was starting to say good bye to mosfets.
I guess I was wrong. Good for me😉
Regards...
Hi Pejinm,
have you fitted some form of electronic limiting?
eg. Zeners across the gate source leads or another current limit?
If the fuses are your only protection then I too am surprised the output stage survived. You were a brave man to check it again!
have you fitted some form of electronic limiting?
eg. Zeners across the gate source leads or another current limit?
If the fuses are your only protection then I too am surprised the output stage survived. You were a brave man to check it again!
pejinm said:
pejinm said:
pejinm,
I suspect your use of 5A fast blow fusing protected the amp is the same as the 5A rating is for 120W into 4 Ohms for 2 FETs at 40V rails. The 40V rails is almost half your rail voltage. I believe you have 6 FETs which in combination with the fast blow fuse of 5A likely allowed the fuse to blow before the FETs had their current/thermal shared limits reached to result in blowing any of the FETs. In other words you had the protection for a 2 FET with 40V rails with your 70V rail 6 FET module netting out to 5A fast blow fusing being very conservative protection for your modules.
Regards,
John L. Males
Willowdale, Ontario
Canada
27 January 2007 22:34
27 January 2007 22:39 Typo corrections
AndrewT said:
keypunch said:
Hi Andrew & John,
I don't have any kind of protection circuit, just classic quasi board with DC detect. I guess that the mosfets were stronger than my 400VA transformer. If I had an 1000VA transformer, I think that amp wouldn't survive.
Regards...
NMOS200 already tested??
Hello Quasi and all others,
Have you already tested your NMOS200 (to-220) Prototype Board?
My parts are already ordered. I am going for the 2sc1845 at the input and irf640n Mosfets.
Still it would be interesting to know that the NMOS200 works (I dont expect anything else🙂 ) reliable.
Loocking forward to build my to220 version.
Black
Hello Quasi and all others,
Have you already tested your NMOS200 (to-220) Prototype Board?
My parts are already ordered. I am going for the 2sc1845 at the input and irf640n Mosfets.
Still it would be interesting to know that the NMOS200 works (I dont expect anything else🙂 ) reliable.
Loocking forward to build my to220 version.
Black
Hello again,
I am just sitting on the internal layout of the amp.
what you can see on the picture is just a brief idea, not very concrete.
It does not show the mains-wirering (would be to messy to schow every single wire), the primary (mains) wires are going to be routed from under the dc-protection board to the soft-start unit and from beyond that board to the mains switch and back.
For the soft-start unit and the dc-Protection I use my own Layouts, but they are very similar, just a bit more to my needs. Thats why I need the secondary AC voltage from the main transformer.
I still beleive there is a better solution. I.e. If I cut the Dc-Protection in two parts, then things would be mouch easier. But thats impossible.
Any Ideas, comments?
Black
I am just sitting on the internal layout of the amp.
what you can see on the picture is just a brief idea, not very concrete.
It does not show the mains-wirering (would be to messy to schow every single wire), the primary (mains) wires are going to be routed from under the dc-protection board to the soft-start unit and from beyond that board to the mains switch and back.
For the soft-start unit and the dc-Protection I use my own Layouts, but they are very similar, just a bit more to my needs. Thats why I need the secondary AC voltage from the main transformer.
I still beleive there is a better solution. I.e. If I cut the Dc-Protection in two parts, then things would be mouch easier. But thats impossible.
Any Ideas, comments?
Black
Attachments
BlacK_Chicken said:
Black,
The layout looks great with one exception. I would try to see if you can change the input path so it does not pass across the power devices. The current wiring path of the inputs presents a possibility for induced noise/signal feedback into the input.
I know you said you cannot split the DC protection in two and that it has some features you like. Could you not elevate the DC Protection in the case and then move the input a bit to the inner part of the rear to run the input leads to the amp modules without crossing over the amp board? Then you could run the module output and the DC Protection output high until they are at the outer area where the input used to be?
I assume you are not using the quasi DC Protection?
Regards,
John L. Males
Willowdale, Ontario
Canada
28 January 2007 19:05
pejinm said:
Pejinm,
I suspect if the only difference in your amp was to have a 1000VA transformer instead of 400VA you would still have the same end result you have expereinced so far. The fuse would still respond in the same time with the same rise in current. The only difference with the 1000VA transformer would be it would heat up a little slower in a short circuit situation at the module outputs. In the end the fuse does not care what the transformer rating is. If it was a 100VA transformer the same result would happen other than the 100VA transfomer would heat up more than the 400VA one before the fuse broke the short condition.
If the transformer was really under dimensioned it would overheat and perhaps melt the copper windings before or as the fuse blows. The only two reasons this would happen is a very under dimensioned transformer for the amp or excessively large fusing. In latter case the amp will never be able to deliver the power expected and would have severe clipping with all the hazzards of clipping to module and speakers. The latter case would in all likelyhood cause the output devices and related front end devices to fry before the fuse could or ever blow.
You happened to have a a fuse dimensioned on the conservative side with respect to the current handling ability of yoru modules which is likely why the output devices have not blown.
That said I would agree with quasi, not to short the output of the modules.
Regards,
John L. Males
Willowdale, Ontario
Canada
28 January 2007 23:29
BlacK_Chicken said:
keypunch said:
I think the layout looks fine Black. With respect to the input leads, I think you have sufficient distance from noisy bits. Just avoid any parallel paths with AC or high current connections. The only thing I would do is swap the position of the slow turn-on PCB with the small transformer, but that's personal preference only and will allow more mains wiring to pass under the board (might look neater.
Cheers
Q
Thank's alot for Input!
Hello,
Your input is much appreciated.
First thing I'll do is to make the PSU-Boards a bit smaller, i.e. to 80 or even 70 x 160 depending on the capacitors.
This was also my first thought. At first I had both RCA jacks nearly on the same hight as the input pads on the NMOS boards, but after a few hours I decided that the jacks are too close to the mains-wires and to close to the Output leads (You know coupling output to input makes the amp instable).
The main probem of this drawing is, thats its only 2-dimensional. In reality I will give the input wires some distance to the board, so they won't lie on the (probably hot) power-devices.
Well, my dc-protection is very similar. First the highpass, then the 3 transistor-arrangement, then a current-source for the relays with BD679/681. The relay coils are wired in series. The hole unit is powered by the main transformer, having the profit of a switch-on delay time of about 5 to 10 seconds. The two 24V relays allow DC rails from 32V to over 60VDC, the use of an external DC- PSU (as Quasi does) is possible.
->"ZMP" is the centertap of the PT.
The Layout:
Ok, good idea. But I need this Board to be near to the primary wires of the big transformer.
As you can see:
The jacks on the board for the Transformer primaries are on the front side (if you look at the drawing of my internal layout), and the AC input for the small transformer are at the rear.
Actually the soft-start Unit is the same as Quasis, with a different relay-type/pinout, LED-out and added Relay to switch the Main transformer.
I already had a fitting switch by hand, but its current-rating was too weak for the 500VA PT, as relays are much cheaper than good quality switches, i decided to go for this (more flexible) solution. Still, if you like to switch both transoformers with one main switch, its easy to place a jumper over the relay-contacts.
I also included the 24V Output. But schamefully a cannot use it for my dc-protection, as I already ordered two 24V relays (as they are connected in series this wont work). I dont want to buy new relays.
Maybe i am way too careful (and mad?) about the internal layout, but ive desinged tube-amp layouts in the past and these are very sensitve soncerning the internal layout and the leaddress (high source impedance, ac heaters, no shielded cables ect.).
I just want to get the best possible performance for
Greetings
Black
Hello,
Your input is much appreciated.
First thing I'll do is to make the PSU-Boards a bit smaller, i.e. to 80 or even 70 x 160 depending on the capacitors.
This was also my first thought. At first I had both RCA jacks nearly on the same hight as the input pads on the NMOS boards, but after a few hours I decided that the jacks are too close to the mains-wires and to close to the Output leads (You know coupling output to input makes the amp instable).
The main probem of this drawing is, thats its only 2-dimensional. In reality I will give the input wires some distance to the board, so they won't lie on the (probably hot) power-devices.
Well, my dc-protection is very similar. First the highpass, then the 3 transistor-arrangement, then a current-source for the relays with BD679/681. The relay coils are wired in series. The hole unit is powered by the main transformer, having the profit of a switch-on delay time of about 5 to 10 seconds. The two 24V relays allow DC rails from 32V to over 60VDC, the use of an external DC- PSU (as Quasi does) is possible.
->"ZMP" is the centertap of the PT.
The Layout:
An externally hosted image should be here but it was not working when we last tested it.
Ok, good idea. But I need this Board to be near to the primary wires of the big transformer.
As you can see:
An externally hosted image should be here but it was not working when we last tested it.
The jacks on the board for the Transformer primaries are on the front side (if you look at the drawing of my internal layout), and the AC input for the small transformer are at the rear.
Actually the soft-start Unit is the same as Quasis, with a different relay-type/pinout, LED-out and added Relay to switch the Main transformer.
I already had a fitting switch by hand, but its current-rating was too weak for the 500VA PT, as relays are much cheaper than good quality switches, i decided to go for this (more flexible) solution. Still, if you like to switch both transoformers with one main switch, its easy to place a jumper over the relay-contacts.
I also included the 24V Output. But schamefully a cannot use it for my dc-protection, as I already ordered two 24V relays (as they are connected in series this wont work). I dont want to buy new relays.
Maybe i am way too careful (and mad?) about the internal layout, but ive desinged tube-amp layouts in the past and these are very sensitve soncerning the internal layout and the leaddress (high source impedance, ac heaters, no shielded cables ect.).
I just want to get the best possible performance for
Greetings
Black
Hello ,
Im happy , my N chanell MOSFET amplifier version TO 247 plastic its ready to testing , but only next week 🙂
Im happy , my N chanell MOSFET amplifier version TO 247 plastic its ready to testing , but only next week 🙂
An externally hosted image should be here but it was not working when we last tested it.
alex mm said:
Alex,
Excellent work.
Did you just place two of the mini quasi boards side by side before you made the PCB?
I am curious how tight the hole you made in the heatsink mounting bracket is? I am not questioning the mounting you did. I trying to gain a sense how tight the hole is or not for the balance between amount of heatsink compound, heatsink and the BC546 Vbe.
I would consider using a TO-92 for the Vbe, but I like flexibility if I need to remove a module for some reason and have yet to figure out a different mechanical way to make the thermal mounting. I know from some commerical designs all they do attach the flat side a TO-92 Vbe to the heaksink with heatsink compound. Not an inspiring method. The hole approach quasi uses makes more sense than sticking with just heatsink compound.
I was not quite able to read the IRF part you used for your output devices. At best these are hard to read when in hand. What IRF device did you choose to use?
Will you be heaksinking the two pre-drivers?
Regards,
John L. Males
Willowdale, Ontario
Canada
31 January 2007 21:10
alex mm said:
Did you have a robot insert those components or are you really just that good? 🙂 It is not very often a diy etch looks this good after component placement. Perhaps I'm getting the shakes as I get older?
Great work! We need to hear how it sounds.
Shawn.
Hi Alex mm
I hope this isn't a dumb question but are the MJE340/350's on the heatsink bracket insulated?
And I just know you're going to put a small heatsink on the second stage transistors...aren't you?
Cheers
Q
I hope this isn't a dumb question but are the MJE340/350's on the heatsink bracket insulated?
And I just know you're going to put a small heatsink on the second stage transistors...aren't you?
Cheers
Q
Hello Alex,
Very nice PCB.
I first thought about the same method, to place two boards on one pcb, but then I changed to two heatsinks and similar two pcbs.
Greetings Black
Very nice PCB.
I first thought about the same method, to place two boards on one pcb, but then I changed to two heatsinks and similar two pcbs.
Greetings Black
... thanks , yes its a great project , i wish to answer to questions first .... to keypunch --- yes i have place two mini quasi boards side by side . The hole in heatsink its 4.5 mm , to mount tight BC 546 i solve with super glue . The heatsink for two pre-drivers , i intend to make it from aluminium foil . Output finals are IRFP 460
... to Tom Waits --- what can i say , 🙂 i am the robot ? just hobby
.... to Quasi thanks again , realy great design , nice devices placement to PCB , yes the MJE 340/350's insulated with mica and siliconic grease .
.... to Black Chicken .. only, my option to place two amps on the same PCB
Sorry for my poor english , i just started to learn more and more
... to Tom Waits --- what can i say , 🙂 i am the robot ? just hobby
.... to Quasi thanks again , realy great design , nice devices placement to PCB , yes the MJE 340/350's insulated with mica and siliconic grease .
.... to Black Chicken .. only, my option to place two amps on the same PCB
An externally hosted image should be here but it was not working when we last tested it.
Sorry for my poor english , i just started to learn more and more
Hi Alex,
Wow, I really, really like what you've done with Quasi's design. Very clean!
Great job there!
-Chris
Edit: After listening to it, replace all those ceramic caps with polystyrene or mica and have another listen. 😉
Wow, I really, really like what you've done with Quasi's design. Very clean!
Great job there!
-Chris
Edit: After listening to it, replace all those ceramic caps with polystyrene or mica and have another listen. 😉
Very nice....when you put that to work ?
Maybe some measurements on the final tests....and after that I think you MUST work on the enclosure with the same passion.
Keep your mind focus on details...that details will give you an impressive good looking amp.
Maybe some measurements on the final tests....and after that I think you MUST work on the enclosure with the same passion.
Keep your mind focus on details...that details will give you an impressive good looking amp.
My finished quasi amp...
I have 2mV DC offset (out-to-gnd) with input conected to GND.
(too bad that when i conect it to computer sound card i get some noise.... but that's my computer fault)
Bias curent 27mV on positive rail and 29mA 😕 on negative rail ... i couldn't set them both equal (why?)
Anyway has a beautiful and clear sound...
I wander ... if I increase the bias curent to 100mA ... would it be possible ? ... would it increase the performance ?
[ Main heatsink is just for test ! 😀 ]
I have 2mV DC offset (out-to-gnd) with input conected to GND.
(too bad that when i conect it to computer sound card i get some noise.... but that's my computer fault)
Bias curent 27mV on positive rail and 29mA 😕 on negative rail ... i couldn't set them both equal (why?)
Anyway has a beautiful and clear sound...
I wander ... if I increase the bias curent to 100mA ... would it be possible ? ... would it increase the performance ?
[ Main heatsink is just for test ! 😀 ]
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
