Email sended.
WARNING : BOARD I SHOWN ARE IN DEVELOPPEMENT AND CAN'T BE TAKEN AS FUNCTIONNALY DEBBUGED VERSION
I shown what i done here to have advices/opinion on it....
Marc
WARNING : BOARD I SHOWN ARE IN DEVELOPPEMENT AND CAN'T BE TAKEN AS FUNCTIONNALY DEBBUGED VERSION
I shown what i done here to have advices/opinion on it....
Marc
oki doki
up and runing
had to debug a couple of mistakes i ve made
data :
63+63 volt rails
6 pairs MJW 21194
arround 200w
the only problem i get is that at idle even if idle is 40 ma the pair MJE 340-350 gets pretty warm ....dangerous warm id say ....i wonder what will happen in full power ...
all is as in the original schematic except MJE 350/TR6 miller cap instead of 39pf is 33 pf since this was available in a good quality
in the pair there is a 3x5cm x2mm piece of alouminum but doesnt look really enough
also at square wave arround 10khz i got my shelf some ringing
have i missed something ????
up and runing
had to debug a couple of mistakes i ve made
data :
63+63 volt rails
6 pairs MJW 21194
arround 200w
the only problem i get is that at idle even if idle is 40 ma the pair MJE 340-350 gets pretty warm ....dangerous warm id say ....i wonder what will happen in full power ...
all is as in the original schematic except MJE 350/TR6 miller cap instead of 39pf is 33 pf since this was available in a good quality
in the pair there is a 3x5cm x2mm piece of alouminum but doesnt look really enough
also at square wave arround 10khz i got my shelf some ringing
have i missed something ????
Stereo SPLs
at full power most listeners expecting high quality will be using an average output of ~2W.
The heatsinks will hardly change in temperature compared to the quiescent condition.
If you are using 87dB/2.8V/m speakers you will be achieving average SPL levels ~93dB @ 1m and ~85dB @ 2.5m.
That is LOUD.
I hope you are not expecting Disco quality.
The drivers are dissipating 3W each.
using a 5C/W sink shared between the two drivers will create Ts~=39 above Ta.
Inside the case Ta could be anywhere between 25degC and 50degC.
5C/W is too small for Ta>=30degC.
Check the temp de-rating for Ta=30degC and P=3W
at full power most listeners expecting high quality will be using an average output of ~2W.
The heatsinks will hardly change in temperature compared to the quiescent condition.
If you are using 87dB/2.8V/m speakers you will be achieving average SPL levels ~93dB @ 1m and ~85dB @ 2.5m.
That is LOUD.
I hope you are not expecting Disco quality.
The drivers are dissipating 3W each.
using a 5C/W sink shared between the two drivers will create Ts~=39 above Ta.
Inside the case Ta could be anywhere between 25degC and 50degC.
5C/W is too small for Ta>=30degC.
Check the temp de-rating for Ta=30degC and P=3W
for fun's sake ....
you know andrew ..... i really like you very much .... i try to follow most of the things you write .Always there something to learn from it . Also all is said ...short ...strict... correct and without any sauce ...
this is nice ...there is a lot of theory and so many nice things said from you in the forum ....
thanks Andrew ....
PS
from a funny point of view
regardless tha above said ....i wish you were closer to meet with the "Greek" side of things ... approach is never that detailed ....too much theory is overlooked ....too many calculations also ....BUT machines are working ....maybe about...may be not correct ....may be stressed ....but working ....
many greeks were emigrants in Germany to work in factory after WW2 arround 1945 so the job was to drill ( properly drill ) 20 heatsinks a day ...( for example ) so the Greek migras was quick quick quick produced actually 40 heatsinks to be on the safe side and then he thought that ""as long as i produced 40 pcs i may excuse my shelf now to the cafeteria """
germans noticed and the greek lost his job ....without a job the Greek mooved again to another place like USA and actualy preserved the same concept about driling heatsinks ...
in usa though he become a manager ..... ha ha ha ha
best regards
you know andrew ..... i really like you very much .... i try to follow most of the things you write .Always there something to learn from it . Also all is said ...short ...strict... correct and without any sauce ...
this is nice ...there is a lot of theory and so many nice things said from you in the forum ....
thanks Andrew ....
PS
from a funny point of view
regardless tha above said ....i wish you were closer to meet with the "Greek" side of things ... approach is never that detailed ....too much theory is overlooked ....too many calculations also ....BUT machines are working ....maybe about...may be not correct ....may be stressed ....but working ....
many greeks were emigrants in Germany to work in factory after WW2 arround 1945 so the job was to drill ( properly drill ) 20 heatsinks a day ...( for example ) so the Greek migras was quick quick quick produced actually 40 heatsinks to be on the safe side and then he thought that ""as long as i produced 40 pcs i may excuse my shelf now to the cafeteria """
germans noticed and the greek lost his job ....without a job the Greek mooved again to another place like USA and actualy preserved the same concept about driling heatsinks ...
in usa though he become a manager ..... ha ha ha ha
best regards
I come from an industry where all our work had to be independently checked.
It could not be signed off by the section head until that was proved by one of our colleagues.
As a Civil Engineer working in the Gas industry we had to consider above all else the Safety of our customers, the public and our employees.
It's the way I grew up through my indenture and beyond. It's normal commercial practice for professionals when Safety is concerned.
Why should my standards and philosophy, that seems so natural, be thrown away? Because it's an inanimate electrical component some of which can kill us if abused?
It could not be signed off by the section head until that was proved by one of our colleagues.
As a Civil Engineer working in the Gas industry we had to consider above all else the Safety of our customers, the public and our employees.
It's the way I grew up through my indenture and beyond. It's normal commercial practice for professionals when Safety is concerned.
Why should my standards and philosophy, that seems so natural, be thrown away? Because it's an inanimate electrical component some of which can kill us if abused?
Re: Stereo SPLs
Oh yeah - plenty loud.
Subjectively speaking, this is comparable to levels measured inside of a moving subway train.
AndrewT said:
Oh yeah - plenty loud.
Subjectively speaking, this is comparable to levels measured inside of a moving subway train.
well not realy
depenting what your requirments are ....
200w for home use is way too much than requiered.... for my taste something that is tops 50w is more than enough
but for PA application even small one ,200 w is just a drop in the ocean ....
anyway ...found all my mistakes in the quasi board ....a very fine amplifier id say ...a prototype board is up and runing ...very nice bass .... midle is ok .... high is not that clear but nobody is going to look at that .... most quitar players are totaly anable to listen to this frequency he he he ...
depenting what your requirments are ....
200w for home use is way too much than requiered.... for my taste something that is tops 50w is more than enough
but for PA application even small one ,200 w is just a drop in the ocean ....
anyway ...found all my mistakes in the quasi board ....a very fine amplifier id say ...a prototype board is up and runing ...very nice bass .... midle is ok .... high is not that clear but nobody is going to look at that .... most quitar players are totaly anable to listen to this frequency he he he ...
Re: well not realy
I know... I do that for a living... 😀
My comment assumed a home listening environment based on Andrew's remark on "disco quality".
I am looking forward to more pictures of your new amp.
sakis said:
I know... I do that for a living... 😀
My comment assumed a home listening environment based on Andrew's remark on "disco quality".
I am looking forward to more pictures of your new amp.
Maybe for "lo-fi" , Most digital material ,while listening even at 5-10w / channel average will demand transient power of hundreds
of watts for accurate reproduction. Even a movie soundtrack that averages 1-5w can have 200w+ peaks during explosions/action scenes.
??? Then why DIY? One can buy semi-engineered OEM from china/japan -ebay , even get auto-level circuitry (compressor) 😀 ,so as to never worry about peaks.
OS
ok ...this done for testing
it is based on indefixe's pcb but features piggy boards with VI limmiters hiden down under ....
it is not 100% debuged yet ita has a few cosmetic issues on the pcb .... it will requiere a bit of better arangement regarding drivers and a bit of space more for for a biger cooler for the pair MJE 340-350
it is working as we speak runs cool smooth scope agrree also so we have our shelfs one more amp .......
particular one is 63+63 rails 6xMJW21194 per board and all the rest as in the schematic ...
at the prize i buy the semis i think that this is really cost effective
i actually have a complete sassi of one altair amp burned beyond repair in the amp boards ( water inside ) but the only problem is that PSU will produce almost 80v ...too much for this amp i think
there you go chief
it is based on indefixe's pcb but features piggy boards with VI limmiters hiden down under ....
it is not 100% debuged yet ita has a few cosmetic issues on the pcb .... it will requiere a bit of better arangement regarding drivers and a bit of space more for for a biger cooler for the pair MJE 340-350
it is working as we speak runs cool smooth scope agrree also so we have our shelfs one more amp .......
particular one is 63+63 rails 6xMJW21194 per board and all the rest as in the schematic ...
at the prize i buy the semis i think that this is really cost effective
i actually have a complete sassi of one altair amp burned beyond repair in the amp boards ( water inside ) but the only problem is that PSU will produce almost 80v ...too much for this amp i think
there you go chief
Attachments
well OS
actually ...this is done cause its really cost effective since semis are so cheap ....
on the other hand expensive material like trafos heatsinks and caps exist in my shop in kilos since i have so many boxes with all inclusive coming from costumers that had this OEm amps chinese made tha prooved to be very crappy ....
though trafos hetasinks and caps or just ok for a thing like that ...
the paricular box and parts belongs to an amp called LIMIT totaly chinese ...class G one which actually was a very fine amp with only just a few problems
---- low quality semis
----Terible "mechano" loose screws and wiring
---- a protection relay that cannot stand the sreaker curent
---- and finaly ceramics all over for everything ....
actually if you have the time to fix all that limit is a very fine amp ....
ostripper said:
actually ...this is done cause its really cost effective since semis are so cheap ....
on the other hand expensive material like trafos heatsinks and caps exist in my shop in kilos since i have so many boxes with all inclusive coming from costumers that had this OEm amps chinese made tha prooved to be very crappy ....
though trafos hetasinks and caps or just ok for a thing like that ...
the paricular box and parts belongs to an amp called LIMIT totaly chinese ...class G one which actually was a very fine amp with only just a few problems
---- low quality semis
----Terible "mechano" loose screws and wiring
---- a protection relay that cannot stand the sreaker curent
---- and finaly ceramics all over for everything ....
actually if you have the time to fix all that limit is a very fine amp ....
Re: oki doki
T7 and T6 will dissipate about 1.5 watts so a bigger heatsink for these is a good idea. But they are very tough transistors so don't worry too much.
You should not be getting any ringing with the square wave test, although the output filter may be contributing. Try measuring before the filter. If you still have ringing then try changing all the 0.1uf bypass caps to a different type or brand (0.1 to 0.47 etc is ok).
The bandwidth of the amp is quite wide so you should have great treble performance. Is the schematic exactly the same as the original or has it been changed in some way?
Anyway good to see its working.
Yia sou
Quasi
sakis said:
T7 and T6 will dissipate about 1.5 watts so a bigger heatsink for these is a good idea. But they are very tough transistors so don't worry too much.
You should not be getting any ringing with the square wave test, although the output filter may be contributing. Try measuring before the filter. If you still have ringing then try changing all the 0.1uf bypass caps to a different type or brand (0.1 to 0.47 etc is ok).
The bandwidth of the amp is quite wide so you should have great treble performance. Is the schematic exactly the same as the original or has it been changed in some way?
Anyway good to see its working.
Yia sou
Quasi
Re: Re: oki doki
Hi quasi, the board is based on a modify early version of mine....based on yours. I juste change TO3 devices to TO264 devices, and adapted placement to parts i have acces. Normally the majors rooting design is based on yours, particulary ground rooting. So The result must be coherant to yours. I made some more cosmetic arrangement to optmise size of board and swap speaker protection outside amp board. I made a sigle pair BJT version based on Nmos200 rooting. I will post later this to work to have your (and others) opinion. Take care that tis work as i wrote it over is not tested and perhaps not completly debbuged.
regards Marc
quasi said:
Hi quasi, the board is based on a modify early version of mine....based on yours. I juste change TO3 devices to TO264 devices, and adapted placement to parts i have acces. Normally the majors rooting design is based on yours, particulary ground rooting. So The result must be coherant to yours. I made some more cosmetic arrangement to optmise size of board and swap speaker protection outside amp board. I made a sigle pair BJT version based on Nmos200 rooting. I will post later this to work to have your (and others) opinion. Take care that tis work as i wrote it over is not tested and perhaps not completly debbuged.
regards Marc
....more results
well thanks quasi for looking in to this ..... the amp is just fine the only thing that was wrong was D1 placed with the wrong polarity ...the heatsink is enough ....
now the drivers are getting a bit hot but also they are not placed in a proper hetasink ( not located in the main heatsink for a number of reasons )
idefixes pcb is perfect ....
to my opinion though some cosmetic changes are to be made ....
----i think just a bit more space arround output transistors will be good
---- more space around the MJE340-350 for a biger heatsink
---- and finally outputs and drivers are not alligned properly ...i think drivers need to go a bit more up
otherwise all is fine ...but more tests will proceed and see how its going
as about the ringing ...after replacing the D1 in the correct polarity is gone .... i got some crisp though in the uper waveform of 1 KHZ sine just before cliping .... after and before cliping all is as supposed to ... weird ...but i ma looking in to it
well thanks quasi for looking in to this ..... the amp is just fine the only thing that was wrong was D1 placed with the wrong polarity ...the heatsink is enough ....
now the drivers are getting a bit hot but also they are not placed in a proper hetasink ( not located in the main heatsink for a number of reasons )
idefixes pcb is perfect ....
to my opinion though some cosmetic changes are to be made ....
----i think just a bit more space arround output transistors will be good
---- more space around the MJE340-350 for a biger heatsink
---- and finally outputs and drivers are not alligned properly ...i think drivers need to go a bit more up
otherwise all is fine ...but more tests will proceed and see how its going
as about the ringing ...after replacing the D1 in the correct polarity is gone .... i got some crisp though in the uper waveform of 1 KHZ sine just before cliping .... after and before cliping all is as supposed to ... weird ...but i ma looking in to it
one thing i forgot
this was my first boards .... im going to change a couple of things mostly regrading cooling and so ....and then others will follow ...( i got one gozillion of npn transistors as we speak )
so in the near feature this amp is going to be pushed like hell since its going to be used in small PA application (stage monitor ,small club , ) and stuff like that
that will be fun to see .... may be others constructed for home use .. .... well ...i am not ...
will see and let you know ...
in my fleet of amplifiers i have only one quasi more its greek made i think that its more than 20 years old ( recently refabrished though ) but from default was a monsterous device any way 8xMJ15025 producing 600w but the limiters will only allow arround 450.... then even the balast resistors are like 0.47R 10w each like the peavey amps..... prooved that it can be pushed to any available abuse you can imagine ...
sound wise is more than ok ( come to thing about it ...it might be a copy of cs800 but it doesnt look like it ...i have to check ....)
thanks leventi !!!! ....it looks like a fine amp so far ....
( any chances to run it in 80+80 volts ????? with as many outs is possible of course )
this was my first boards .... im going to change a couple of things mostly regrading cooling and so ....and then others will follow ...( i got one gozillion of npn transistors as we speak )
so in the near feature this amp is going to be pushed like hell since its going to be used in small PA application (stage monitor ,small club , ) and stuff like that
that will be fun to see .... may be others constructed for home use .. .... well ...i am not ...
will see and let you know ...
in my fleet of amplifiers i have only one quasi more its greek made i think that its more than 20 years old ( recently refabrished though ) but from default was a monsterous device any way 8xMJ15025 producing 600w but the limiters will only allow arround 450.... then even the balast resistors are like 0.47R 10w each like the peavey amps..... prooved that it can be pushed to any available abuse you can imagine ...
sound wise is more than ok ( come to thing about it ...it might be a copy of cs800 but it doesnt look like it ...i have to check ....)
thanks leventi !!!! ....it looks like a fine amp so far ....
( any chances to run it in 80+80 volts ????? with as many outs is possible of course )
Here is the latest version
Output transistor and driver are direct screw on heatsink without bracket. The board will be cut on driver zone.
Marc
Output transistor and driver are direct screw on heatsink without bracket. The board will be cut on driver zone.
Marc
Here is the Nbip_100 version based on Nmos 200 with one pair of MJL4182 and MJE15030/15031 for driver.
Marc
Marc
The goal is to realise to amplifier for HT use :
- one with 2 NBip_300 chanel (500Va 2x30Vac) => R and L front chanels
- one with 3 chanels (1 Nbip_300 for center and 2 NBip_100 for rear R and L suround). 660Va 2x30Vac for the 3 chanels.
I have to look how to change value of to resistance to maintain gain and current level for input stage.
Marc
- one with 2 NBip_300 chanel (500Va 2x30Vac) => R and L front chanels
- one with 3 chanels (1 Nbip_300 for center and 2 NBip_100 for rear R and L suround). 660Va 2x30Vac for the 3 chanels.
I have to look how to change value of to resistance to maintain gain and current level for input stage.
Marc
idefixes ....
i think tha yor latest versionj of pcb is runing to the same mechanical error like all the previous ones ....
Drivers MJ15030-31 will require a "large" heatsink , in the original brothe of quasi drawing the drivers +vbe multipier is located in the main heatsink in the center of the all thing so the heatsink will have an easier job to cool drivers down .
locating the drivers so low will need another heatsink ...since the main heatsink ....or a thick corner aluminioum that somehow drivers are placed on and then the small heatsink is thermaly coupled to the main heatsink
let us not forget that also there we need to have a vbe multiplier that needs to be functionable also ....
cuting the board has to be factory made otherwise will look to ugly ( that is not cost efective )
also the all think is not versatile this way since only massive heatsinks can be used and any corner brackets is simply out of the question
just an opinion
Idefixes said:Here is the latest version
Output transistor and driver are direct screw on heatsink without bracket. The board will be cut on driver zone.
Marc
i think tha yor latest versionj of pcb is runing to the same mechanical error like all the previous ones ....
Drivers MJ15030-31 will require a "large" heatsink , in the original brothe of quasi drawing the drivers +vbe multipier is located in the main heatsink in the center of the all thing so the heatsink will have an easier job to cool drivers down .
locating the drivers so low will need another heatsink ...since the main heatsink ....or a thick corner aluminioum that somehow drivers are placed on and then the small heatsink is thermaly coupled to the main heatsink
let us not forget that also there we need to have a vbe multiplier that needs to be functionable also ....
cuting the board has to be factory made otherwise will look to ugly ( that is not cost efective )
also the all think is not versatile this way since only massive heatsinks can be used and any corner brackets is simply out of the question
just an opinion
you even
have to cut the insulation pads .... to fit them in to transitors .... too close to each other ....some mm more to each needed i think
have to cut the insulation pads .... to fit them in to transitors .... too close to each other ....some mm more to each needed i think
- Status
- Not open for further replies.