I'm sure it sounds good... it's ended up very like the "blameless class b" of Doug Self.
Q1 and Q2 ????? 100MA devices... that's asking for trouble... and the result would be burned out speakers.
Edit... what is your thinking behind C8 and C12 ?
Q1 and Q2 ????? 100MA devices... that's asking for trouble... and the result would be burned out speakers.
Edit... what is your thinking behind C8 and C12 ?
Last edited:
Without c8 and c12 the amp oscillate's.
it limits the current source
And the bc107c and bc177c can handle 200Ma collector current and they are cooled so they can handle 0.65w collector dissipation.
altough the 2sc1826 and 2sa768 are relatively low power devices they have a hfe between 40 and 400 so the are easy to drive and they are fast 10MHZ transistion frequency.
i uploaded thedatasheets in the pervious posts.
Greets captain archer out.............!!!
it limits the current source
And the bc107c and bc177c can handle 200Ma collector current and they are cooled so they can handle 0.65w collector dissipation.
altough the 2sc1826 and 2sa768 are relatively low power devices they have a hfe between 40 and 400 so the are easy to drive and they are fast 10MHZ transistion frequency.
i uploaded thedatasheets in the pervious posts.
Greets captain archer out.............!!!
Hi,
I'm not trying to sound critical... and I would hate to see you damage or destroy a good pair of speakers because of this.
The BC107 is 100ma collector current (maximum continuos). 200ma is specified as a peak current... my data doesn't seem to specify the time duration, but I'm sure anyone reading this will confirm that they are not robust enough for drivers.
At full output into say 4 ohms have you simulated (or measured) the base drive requirements of the output devices. Remember that current gain in the outputs will drop off at higher levels of collector current (beta droop).
I'm not trying to sound critical... and I would hate to see you damage or destroy a good pair of speakers because of this.
The BC107 is 100ma collector current (maximum continuos). 200ma is specified as a peak current... my data doesn't seem to specify the time duration, but I'm sure anyone reading this will confirm that they are not robust enough for drivers.
At full output into say 4 ohms have you simulated (or measured) the base drive requirements of the output devices. Remember that current gain in the outputs will drop off at higher levels of collector current (beta droop).
bc107 is a 45Vce0 and 100mAIc device.
It could act as a driver to a 1 or 2W headphone amplifier.
It is unsuitable, as driver, for an 8ohm speaker amplifier.
C8 & C12 when fitted usually connect between supply rail and base, not collector and base.
It could act as a driver to a 1 or 2W headphone amplifier.
It is unsuitable, as driver, for an 8ohm speaker amplifier.
C8 & C12 when fitted usually connect between supply rail and base, not collector and base.
2sa768 is a 60Vce0 4AIc 30W device.
This will struggle to drive an 8ohm speaker reliably.
Note that hFE min for Ic=1A is 40.
I cannot find what hFE min drops to for Ic=2A, nor 3A, nor 4A.
This will struggle to drive an 8ohm speaker reliably.
Note that hFE min for Ic=1A is 40.
I cannot find what hFE min drops to for Ic=2A, nor 3A, nor 4A.
It does... top of page 2. 100ma MAX.
I'm going to add something else now... I looked at your data for the outputs... and 4 amp maximum collector current again isn't really enough. You have to imagine all the worst case scenarios, switch on surges, switching on with the volume turned fully up... a momentary short on the outputs, and of course a reactive speaker load. Transistors suffer from an effect called secondary breakdown where they fail suddenly even though the ratings might suggest otherwise. This usually happens at collector currents near to, but below the maximum rating and at highish VCE voltage... it's a well known effect.
Your amp has no SOAR or safe operating area protection to guard against this.
Don't take it as a critisism... try and understand why they are not really suitable.
Yes for lowish domestic listening you will probably get away with it... but take the design further... try and anticipate all that could happen and design against that possibility.
I'm going to add something else now... I looked at your data for the outputs... and 4 amp maximum collector current again isn't really enough. You have to imagine all the worst case scenarios, switch on surges, switching on with the volume turned fully up... a momentary short on the outputs, and of course a reactive speaker load. Transistors suffer from an effect called secondary breakdown where they fail suddenly even though the ratings might suggest otherwise. This usually happens at collector currents near to, but below the maximum rating and at highish VCE voltage... it's a well known effect.
Your amp has no SOAR or safe operating area protection to guard against this.
Don't take it as a critisism... try and understand why they are not really suitable.
Yes for lowish domestic listening you will probably get away with it... but take the design further... try and anticipate all that could happen and design against that possibility.
using a +-20Vdc supply will take the drivers to a maximum 40V + mains and transformer voltage tolerance. That's about as high as you dare take a 45Vce0 device.
Using an 8ohm speaker from +-20Vdc requires a peak resistive current of ~2.2Apk.
Into a real speaker load allow for 3times this or ~ 7Apk on short term transients.
If the hFE of the a768 remains at 40 then the worst case base current would be ~170mApk and this must be supplied by the driver.
It is more likely that worst case current demand from the driver could be as high as 300mApk for a reactive 8ohm speaker.
a768 and bc107 do not work for an 8ohm capable power amp on +-20Vdc supply rails.
if we get 18Vpk into an 8r0 load then maximum power is ~20W.
Typically you would need a pair of ~60W devices to survive this duty.
Using an 8ohm speaker from +-20Vdc requires a peak resistive current of ~2.2Apk.
Into a real speaker load allow for 3times this or ~ 7Apk on short term transients.
If the hFE of the a768 remains at 40 then the worst case base current would be ~170mApk and this must be supplied by the driver.
It is more likely that worst case current demand from the driver could be as high as 300mApk for a reactive 8ohm speaker.
a768 and bc107 do not work for an 8ohm capable power amp on +-20Vdc supply rails.
if we get 18Vpk into an 8r0 load then maximum power is ~20W.
Typically you would need a pair of ~60W devices to survive this duty.
Last edited:
i agree on the output transistors but again but this a 15 watt amp max and it is designed for this goal see all other components.
so the devices will suffice.
Funny though that the output transistors came from a so called 2x20 watt comercial Erres amp you know the big looks but little special inside🙄
they where especialy well cooled though.😀
i understand what you guys are trying to suggest and i agree on the Safe operating area but the transformer driving the unit will be 75va at 2x15v so
only the surges from short circuits or 4ohm speakers at full power will be a problem but so it was for the sony and the erres amp. and they work till this day (well exept the erres now😀)
the sony amp incorporates also 4a collector current devices and they are rated at 40 watts max but with a lower Beta so...
but whats the deal with the cdil datasheets then???😕
hmm strange but i did use cdil devices so i am going to stick with the cdil datasheets.
Also strange, the cdil datasheets say that the bc177 can handle even a little bit more collector disipation but the Philips datsheets give the same specs for bc107 and bc177😕
ok i have tested the amp at full power to clipping now for the last 5 minutes while typing and all small signal transistors are slightly warm and output warm to hot, well it is only cooled to a little L shaped aluminium bar now, this will be mounted to a metal chassis.
captain out...................!!
so the devices will suffice.
Funny though that the output transistors came from a so called 2x20 watt comercial Erres amp you know the big looks but little special inside🙄
they where especialy well cooled though.😀
i understand what you guys are trying to suggest and i agree on the Safe operating area but the transformer driving the unit will be 75va at 2x15v so
only the surges from short circuits or 4ohm speakers at full power will be a problem but so it was for the sony and the erres amp. and they work till this day (well exept the erres now😀)
the sony amp incorporates also 4a collector current devices and they are rated at 40 watts max but with a lower Beta so...
but whats the deal with the cdil datasheets then???😕
hmm strange but i did use cdil devices so i am going to stick with the cdil datasheets.
Also strange, the cdil datasheets say that the bc177 can handle even a little bit more collector disipation but the Philips datsheets give the same specs for bc107 and bc177😕
ok i have tested the amp at full power to clipping now for the last 5 minutes while typing and all small signal transistors are slightly warm and output warm to hot, well it is only cooled to a little L shaped aluminium bar now, this will be mounted to a metal chassis.
captain out...................!!
testing a power amp into a resistive load is not nearly as stressfull as driving the same nominal impedance speaker.
I used to assess my amplifiers' current capability by checking into a resistor of half nominal load impedance.
I have now changed to looking at SOAR with resistor load = Nominal Impedance/3
Check to see if the drivers are both 8ohm. The bass/mid might be a 6ohm or 4ohm driver.
I used to assess my amplifiers' current capability by checking into a resistor of half nominal load impedance.
I have now changed to looking at SOAR with resistor load = Nominal Impedance/3
Check to see if the drivers are both 8ohm. The bass/mid might be a 6ohm or 4ohm driver.
i guess i posted yust before you well i tested the unit into 8 ohm Kef chorale speakers at full power for about ten minutes now
By the way now i hav a good question for my boss (i work at Philips)😛
see wat hee has to say about this😀
well it works fine so if i'l blow it up i will rebuild it but i think it wont happen.
it has been playing realy loud now i am getting deaf now so i'm turning the volume down now.
greets captain out....!!!
By the way now i hav a good question for my boss (i work at Philips)😛
see wat hee has to say about this😀
well it works fine so if i'l blow it up i will rebuild it but i think it wont happen.
it has been playing realy loud now i am getting deaf now so i'm turning the volume down now.
greets captain out....!!!
and "fullpower" was probably an average of 500mW with regular peaks hitting 5W. Just occasional transients getting clipped, so the sound was not too gritty and spity and sounding too loud..
Did any components get warm or hot?
no, and i am testing the amp for days now at full power into a normal 8Ohm speaker load, and the only transistor that gets slightly warm is the BC177 in the vas current source.
But i measured the temerature of this device and it was 49* celcius at max power output for 2 hours.
tested it with all kinds of music from beethoven to basshunter and i am driving people in my house nuts with the music so loud so 500mw average
hmmmm well....................!
again i agree on the safe operating area but the statement that these devices are not capable of driving a 8 Ohm load is a little over exaggerated since the original Sony Ta-70 amplifier used the 2SD292 devices and they where 3 amp's collector current and 18 watts power dissipation and these transistors where also driven by even lower power to92 drivers.
And the amp was rated at 2x15w and survived until this day without servicing.
And again the transistors where salvaged from an Erres 2x20 watt amp.
and like the datasheet says (for use in low frequency power audio amplifier applications)
and these output transistors where driven by some 100ma PEAK to92 2sa and 2scxxxx devices.
and then the Sony TA-1630, it uses almost the same output devices and also works fine until this day and i think i am like the 154'th owner or something so.
included a photo of the erres amp
Attachments
- Status
- Not open for further replies.