50W amp new schematic
Hi djk actually i added the capacitor in the ccs for the LTP because the amp was having an audible hum and upon insertion of the cap the hum dissapeared. I've seen the same cure used in one of JLH's design. Maybe the collector resistor of the first LTP changed from 2.2k to 4.7k could cure the problem i haven't tried yet. Otherwise are there any additional comments you might think of regarding the last schematic i posted?
Thanks
Hi djk actually i added the capacitor in the ccs for the LTP because the amp was having an audible hum and upon insertion of the cap the hum dissapeared. I've seen the same cure used in one of JLH's design. Maybe the collector resistor of the first LTP changed from 2.2k to 4.7k could cure the problem i haven't tried yet. Otherwise are there any additional comments you might think of regarding the last schematic i posted?
Thanks
Hi zeus_threat
Your amp looks a lot better now. But it's like a state-of-the-art from 10 years ago!
Why R20=100 ohms?
In a Miller compensated amp you don't need 100 ohms but 10 ohms might be useful, or pick a value you can use as a current limiter in case of overloading. BD140's have pretty flat output characteristics and should not give much Early voltage distortion.
You might want to consider using a current mirror instead of the resistor load on the base of the VAS like Self's blameless.
I often use R20=100 ohms BTW but not in Miller compensated amplifiers. It reduces the overall gain but increases the frequency response.
cheers
John
Your amp looks a lot better now. But it's like a state-of-the-art from 10 years ago!
Why R20=100 ohms?
In a Miller compensated amp you don't need 100 ohms but 10 ohms might be useful, or pick a value you can use as a current limiter in case of overloading. BD140's have pretty flat output characteristics and should not give much Early voltage distortion.
You might want to consider using a current mirror instead of the resistor load on the base of the VAS like Self's blameless.
I often use R20=100 ohms BTW but not in Miller compensated amplifiers. It reduces the overall gain but increases the frequency response.
cheers
John
Hum problem
Hi john, my amp does not feel fresh anymore John. joking anyway. I'll reduce that value to 39Ohm DJK. I built a version of the amp on strip board and tested it with a 15-0-15 supply with low volume. it sounds geat. But there is a soft hum coming out of the amp. Could it be due to the first CCS? i chekced all wiring everything is correct
Thanks
Zeus
Hi john, my amp does not feel fresh anymore John. joking anyway. I'll reduce that value to 39Ohm DJK. I built a version of the amp on strip board and tested it with a 15-0-15 supply with low volume. it sounds geat. But there is a soft hum coming out of the amp. Could it be due to the first CCS? i chekced all wiring everything is correct
Thanks
Zeus
Hi zeus_threat
There are two most likely possibilities I suspect. The first is the one you've already mentioned. I posted my favourite CCS earlier - will find the thread/post if you want. It has a zener diode to stop ripple from the power lines reaching the CCS and a second zener to define the current.
It beats having capacitors which cause turn-on thumps.
THe other possibility is in your grounding. Much has been said in the grounding threads (two of them appear to be on the go) and it seems the best bet is to connect the inputs on isolated RCA phono sockets to the amp; take a ground lead to CEP (common earth point or star) from the input stage and a separate power lead ground to the CEP.
The chassis needs the earth isolation resistor. Best bet is Bob Cordell's resistor plus power diodes in case of large earth fault currents.
cheers
John
There are two most likely possibilities I suspect. The first is the one you've already mentioned. I posted my favourite CCS earlier - will find the thread/post if you want. It has a zener diode to stop ripple from the power lines reaching the CCS and a second zener to define the current.
It beats having capacitors which cause turn-on thumps.
THe other possibility is in your grounding. Much has been said in the grounding threads (two of them appear to be on the go) and it seems the best bet is to connect the inputs on isolated RCA phono sockets to the amp; take a ground lead to CEP (common earth point or star) from the input stage and a separate power lead ground to the CEP.
The chassis needs the earth isolation resistor. Best bet is Bob Cordell's resistor plus power diodes in case of large earth fault currents.
cheers
John
hi zeus_threat
my preference for CCS is in thread Randy Sloan's "fig. 11.4" started by tcpip - post 40.
Some discussion of whether a capacitor can be used for bypassing ripple voltages in that thread too.
The advantage of a zener CCS (or a JFET suggested by lineup) is that these don't need capacitors and this means they work fast from turn-on.
Using a red LED gives almost ideal current source - because the change in voltage with temperature almost exactly balances the Vbe change, but a zener voltage operates at quite a high level (5.6V) which is almost as stable.
Don't know whether zener or redLED give more noise. Possibly FET has lower noise but more variability due to differences in Vt.
cheers
John
my preference for CCS is in thread Randy Sloan's "fig. 11.4" started by tcpip - post 40.
Some discussion of whether a capacitor can be used for bypassing ripple voltages in that thread too.
The advantage of a zener CCS (or a JFET suggested by lineup) is that these don't need capacitors and this means they work fast from turn-on.
Using a red LED gives almost ideal current source - because the change in voltage with temperature almost exactly balances the Vbe change, but a zener voltage operates at quite a high level (5.6V) which is almost as stable.
Don't know whether zener or redLED give more noise. Possibly FET has lower noise but more variability due to differences in Vt.
cheers
John
John_Ellis, please make contact with me, there's an old friend that want to have
your E-mail adress.
Please contact me:
nanabrother@yahoo.com
Thanks in advance.
regards,
Carlos
your E-mail adress.
Please contact me:
nanabrother@yahoo.com
Thanks in advance.
regards,
Carlos
Hi zeus-threat
they both show the current in Q1!
you've already got red circles round the "meter" showing current in Q1.
Don't think you need to try too hard to balance the current, but think that the currents could be a little higher.
With temperature change, the PNP next stage will change the balance anyway.
cheers
John
they both show the current in Q1!
you've already got red circles round the "meter" showing current in Q1.
Don't think you need to try too hard to balance the current, but think that the currents could be a little higher.
With temperature change, the PNP next stage will change the balance anyway.
cheers
John
Hum problem
Hi john thanks for the reply. Actually i measured the voltage present at the amp output and it was 50mv dc.
I tested the amp with a +-15V psu and the voltage did not collapse a lot upon application to the amp.
Music signals pass through without any problem
Idling current is 50mA.
Do you have any idea based on these measurements where the hum could be coming from?
Thanks
Patrice
Hi john thanks for the reply. Actually i measured the voltage present at the amp output and it was 50mv dc.
I tested the amp with a +-15V psu and the voltage did not collapse a lot upon application to the amp.
Music signals pass through without any problem
Idling current is 50mA.
Do you have any idea based on these measurements where the hum could be coming from?
Thanks
Patrice
Hi,
I think the current in the LTP emitters is the parameter that should be balanced. It can be measured across the emitter resistors.
When in balance the voltage across both Re are the same and then both emitters are at the SAME voltage. When that holds true then the Vbe of both transistors will be the same WHEN the inverting input voltage = non-inverting input voltage = zero output voltage.
I think that last bit is important so I'll re-phrase it.
When the difference between inverting and non-inverting input voltages are zero, the output voltage will be zero IF the emitter voltages are the same (assuming matched Vbe at quiescent current).
The difference between 272mA and 367mA (95uA) is the current coming from Q4 and that is Q4's [quiescent current/hFE]. A higher gain VAS will reduce this current fed back into the LTP. That higher gain has the effect of letting R5 define the collector load more closely. That 95uA indicates a high VAS current AND a low gain VAS transistor. I would be aiming for <=30uA (6mA/200).
I think the current in the LTP emitters is the parameter that should be balanced. It can be measured across the emitter resistors.
When in balance the voltage across both Re are the same and then both emitters are at the SAME voltage. When that holds true then the Vbe of both transistors will be the same WHEN the inverting input voltage = non-inverting input voltage = zero output voltage.
I think that last bit is important so I'll re-phrase it.
When the difference between inverting and non-inverting input voltages are zero, the output voltage will be zero IF the emitter voltages are the same (assuming matched Vbe at quiescent current).
The difference between 272mA and 367mA (95uA) is the current coming from Q4 and that is Q4's [quiescent current/hFE]. A higher gain VAS will reduce this current fed back into the LTP. That higher gain has the effect of letting R5 define the collector load more closely. That 95uA indicates a high VAS current AND a low gain VAS transistor. I would be aiming for <=30uA (6mA/200).
Hi zeus-threat
start again - pressed return accidentally ...
if you have not got some sort of filter on your CCS hum could be getting into the input stage from the power rails.
I use a zener to filter ripple for this purpose, but many circuits in these threads have shown capacitors taken from a tap point (resistor from - rail, capacitor to ground and capacitor /resistor junction to CCS). IN my view, capacitors always add delay and speaker thump, but either solution might reduce your hum.
Single ended circuits like this have lower power supply rejection than current mirror types and need power rail filtering.
If your amplifier is not oscillating, maybe you have an earth loop between preamp and power amp. I don't want to advise on earthing arrangements because it could be dangerous if done wrongly. But there are some dicussions in these threads. Try searching through these to find recommendations on "earth disconnectors" which stop circulating earthing currents but allow for high current fault flow also.
cheers
John
start again - pressed return accidentally ...
if you have not got some sort of filter on your CCS hum could be getting into the input stage from the power rails.
I use a zener to filter ripple for this purpose, but many circuits in these threads have shown capacitors taken from a tap point (resistor from - rail, capacitor to ground and capacitor /resistor junction to CCS). IN my view, capacitors always add delay and speaker thump, but either solution might reduce your hum.
Single ended circuits like this have lower power supply rejection than current mirror types and need power rail filtering.
If your amplifier is not oscillating, maybe you have an earth loop between preamp and power amp. I don't want to advise on earthing arrangements because it could be dangerous if done wrongly. But there are some dicussions in these threads. Try searching through these to find recommendations on "earth disconnectors" which stop circulating earthing currents but allow for high current fault flow also.
cheers
John
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.