Dear Quasi!
What about using a single transistor (let's say a BD139, coded as T4') instead D3 and T4? Collector of T4' then goes to base of T6, base of T4' then goes to emmiter of T6, and emmiter of T4' then goes to + rails. T4' then would be mounted near the output devices or thermal contact.
I think this way would be more elegant.
Have you tried not to use higher + rails to supply the driver circut? You cannot reach enough output in the + output peaks that way?
So what about using +/-80-100V for the both amp?
What is the purpose of T7 and T9 in your design?
And what is the purpose of R7?
And have you tried the protection? So have you tried to short circut the amp? Is D5 and D4 is enough to protect the output FETs?
T5's Uce is double of T8's Uce. Is it causing any unlinearity or distortions?
Thanks, regards
What about using a single transistor (let's say a BD139, coded as T4') instead D3 and T4? Collector of T4' then goes to base of T6, base of T4' then goes to emmiter of T6, and emmiter of T4' then goes to + rails. T4' then would be mounted near the output devices or thermal contact.
I think this way would be more elegant.
Have you tried not to use higher + rails to supply the driver circut? You cannot reach enough output in the + output peaks that way?
So what about using +/-80-100V for the both amp?
What is the purpose of T7 and T9 in your design?
And what is the purpose of R7?
And have you tried the protection? So have you tried to short circut the amp? Is D5 and D4 is enough to protect the output FETs?
T5's Uce is double of T8's Uce. Is it causing any unlinearity or distortions?
Thanks, regards
hienrich said:hi Quasi: is your beast roaring now.....
actually I'm interested with this one and may I know about
sonic output....
pls. include bias currents on the ouput stage
rgds,
hienrich
Yes this beast has been roaring for some time now and is used and abused as a party system. I run it into some PA drivers i.e high efficiency bass drivers with a piezo mid-high top.
How does it sound? Well at partys it sounds awesome and loud and clear etc....but not audiophile quality I'm afraid. The amp that I built is slightly noisy, which is no problem at partys but can be heard when all is quiet.
I have posted another amp in "power amp under development" which is IMHO "superlative". This amp is not as efficient (for reasons addressed in that thread) but the sonic performance is exceptional (again IMHO & others).
As far as biasing goes, I run about 40mA per FET pair so for 5 pairs = 200mA. You could run anything up to around 100mA per pair but heat will be a problem, around 18 watts per pair in this instance.
Cheers
quasi said:
I am not sure if T7/T9 are necessary. To me, they are nothing but DC shifting devices to make sure that T5/T8 work in roughly the same Vce. the downside of having T7/T9 is that you then are forced to either a) accept rail losses or b) use higher rail for the VAS section.
I would go away with T7/T9.
Also, i would use a MOSFET in place of T4 to have similar thermal characteristics as the output stage.
tlf9999 said:
I have posted a low power version of this design that uses MJE350's instead of the BC327's in "another quasi-complimentary design" thread but there are 2 main reasons for T7 & T9 in this higher power version.
1. To remove excessive voltage from T5 & T8. There is 18 mA running through each of these and if T7 & T9 were not there then T8 would have around 85 volts across it (1.53 watts) and T5 would have 170 volts across it (3.08 watts). These devices would then have to be something different (MJE350) and these would run very hot.
2. By leaving only a few volts across T5 & T8 the second stage is faster and more linear.
The reason for the seperate rail is to lift the available voltage drive for the output stage by 6 volts or so. Without it the amp would clip on the positive waveform before the negative cycle clips. This is not due to the amps topology but the requirement for the output FETs to have at least 4 volts gate to source before they turn on. Under power conditions this requirement exceeds 6 volts. I know I am wasting some voltage in the input stage (across the 4k7 resistors) but once I added the seperate rail I chose to leave more voltage for the second stage constant current source.
You can't really use a mosfet for T4 because this is actually wired up as a diode. In practice I have found that the location of this is more important than it's tracking characteristics. Once the amp settles ( a few minutes) the bias does not vary by more than 20% or so, and for a PA amp this is quite acceptable.
Cheers
Hey Hienrich,
The amp has a bit of hiss and a barely audible hum. It could be the case wiring that is the problem, but I have never taken the time to fix it. This is mainly because the other amp that I posted in "power amp under development" is so good that efforts trying to improve this one seem of less value. VR1 let me assure you adds no (nothing audible or measurable) noise. In fact it helps reduce the overall noise by allowing the output to be zeroed precisely.
But hey as a party system it's awesome and I have had many purchase offers....but I can't sell my babies.....you know I just can't. Sigh!
Cheers
The amp has a bit of hiss and a barely audible hum. It could be the case wiring that is the problem, but I have never taken the time to fix it. This is mainly because the other amp that I posted in "power amp under development" is so good that efforts trying to improve this one seem of less value. VR1 let me assure you adds no (nothing audible or measurable) noise. In fact it helps reduce the overall noise by allowing the output to be zeroed precisely.
But hey as a party system it's awesome and I have had many purchase offers....but I can't sell my babies.....you know I just can't. Sigh!
Cheers
Quasi,
I have the schemes and the PCB lay-out of your early design
which you posted on the other thread.
do you have an upgrade of that?
well, upgraded or not I wish I could find time to look for a better supplier for its components,
locally I couldn't find them..
I will just post it if ever it's done...
meanwhile I will just enjoy myself with BJT's just like this:
I have the schemes and the PCB lay-out of your early design
which you posted on the other thread.
do you have an upgrade of that?
well, upgraded or not I wish I could find time to look for a better supplier for its components,
locally I couldn't find them..
I will just post it if ever it's done...
meanwhile I will just enjoy myself with BJT's just like this:
edl said:
Hey edl, you posted while I was typing.
I tried to setup a normal constant current source with 0.6 volts as a reference for T6. I could have used two diodes on the base here, but I used a transistor (T4) as a diode because it's easier to mount and thermally tracks better. I gues you could use any constant current source here provided there was thermal tracking for the output stage.
I think I answered the reasons for T7 & T9 and the higher rails in my last post.
Without R7 there would be over 90 volts across T2. With R7, T2 only has around 60v....ooops maybe R7 should be 27K, anyway you get the picture.
Cheers
I can't quite figure out the value of Resistor 120, in series with the output, could you please tell me what it it?
A'm sure there are a lot of 1KW+ amps out there, but my preference for redundant lower powers stems from the prices I have to pay for speakers.
I can get good quality 100-200W spaekers for under $60, but as soon as I go higher in power, the prices escalate upwards out of control. My other life responsibilities force electronics to be relegated to a "spend-little-or-nothing" policy.
BTW, has anyone seen any amp designs using IGBTs?
Adrian
hienrich said:
Umm ...which thread do you mean? If you mean "power amp under development" then that board is complete with several built systems in service with some very happy constructors.
If you mean "another quasi-complimentary design" then I have built that amp but not on the PCB layout posted. I have two boards half built only, so I cannot comment on it.
I like your quasi-comp design. Why do you use MJE15033 in your ccs and other stages? I would have thought smaller trasnistors here except of course in the driver stages. I have tried using cascoded input stage myself but have found these to be a bit noisy.
Each to his own I suppose ....nice design though...more efficient than mine too.
Oops edl ...about the Vce on T5 & T8. No real problems her with linearity etc, T7 & T9 help to resolve that.
Cheers
Cheers [/B][/QUOTE]
Quasi : I mean the "power amp under development" thread...
Actually I've used MJE15033 for CCS because of availability...
I have some couple of this just in time when I was soldering all the
components. In my first schemes which I didn't post they were
D669's but I was also thinking about MJE15033's higher breakdown
voltage. Talking about noise, nothing is audible....
not even thumps when turning on the amp......
This amp was made also through the help of the forum ...
rgds,
hienrich
[/QUOTE]
Quasi : I mean the "power amp under development" thread...
Actually I've used MJE15033 for CCS because of availability...
I have some couple of this just in time when I was soldering all the
components. In my first schemes which I didn't post they were
D669's but I was also thinking about MJE15033's higher breakdown
voltage. Talking about noise, nothing is audible....
not even thumps when turning on the amp......
This amp was made also through the help of the forum ...
rgds,
hienrich
tlf9999 said:
Yeah ...I meant with 0.6v as a reference for the ccs transistor. Anyway a FET would work with changes to the emmitter resistor.
Hey Hienrich,
The amp in "power amp under development" can be built using the last posted schematics and board layouts. A few have been built locally plus I believe a few elsewhere.
The local versions perform very well, are absolutely quiet with exceptional sonic performance (IMHO and others). If you make one, please let me know how it goes.
Cheers
Hey Hienrich,
you are running at a voltage gain of about 66, have you tried running it at lower gain of about 20? you may see some improvement in the high end.
btw, have you visited www.pinoydvd.com? it is a local site just like this one.
you are running at a voltage gain of about 66, have you tried running it at lower gain of about 20? you may see some improvement in the high end.
btw, have you visited www.pinoydvd.com? it is a local site just like this one.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.