NJW0281G / NJW0302G instead of 2SC5200/2SA1943

Status
Not open for further replies.
Jump to Latest
#21
Apex14 = baxandall based. While I have even seen a AX14 run perfect at 300W
(still4given), on one of my output stages , unless everything is perfectly sourced ....
this topology WILL oscillate if all is not "right on".
One offspec semi will "excite" the baxandall , I always recommend double checking
the BCxxx's .... no flux , and other extra precautions with this setup.
The driver and output stage choices also count heavily in this amps ultimate stability -
(super-pairs are different , especially as VAS).

OS.
 
#22
Alan0354 said:
I just looked at the Apex AX14, it has bias adjust!!

I don't think changing the transistor will fix your problem. You better find out what is the bottle neck of the sound quality. there are so many factor that affect the sound quality and power amp is only one of the many factors.

I remember when I had the Acurus amp ( so so in audiophile) and a pair of Kef floor standing UniQ ( less than so so!!). It worked really well for the money. Then I went and got a pair of JM LAB 913 series, I hooked it up, it was very disappointing. I was so worry. Then I experiment with the speaker cable. I start pairing up the 12 gauge monster like cable. Every pair I added improve the sound. The sound just "popped". The improvement only stopped after the 4th pair of cable per side. I use 2 pair in parallel for the mid and tweeter, two pairs for the woofer. Then adding the 5th pair does not seems to make any difference.

I can tell you, when I just switch the Kef the JM Lab without changing the cable, there is hardly any improvement. Now, it is day and night difference. The sound becomes 3D with transparency and depth, it just comes alive. The improvement is astonishing.

I still using the cheap Harmon Kardon receiver as preamp. I tried an expensive preamp and it did not sound as good. So I stay with the HK all these years.

My point is trouble shoot your problem first. Your room, move the speaker around. try a different music source, a different DVD/CD player, try a pair of better RCA cable. they do make a difference. I moved to a bigger house, somehow, the sound is not as good as the old house. Maybe it's the hardwood floor. The better the system, the more critical everything is, even the surrounding. There is a good reason why they have hifi furniture, a foam roll, panel etal. the higher the quality, things that never a problem becomes one.

Then you might have something wrong with the power amp also. Don't blame on the power transistor, that's the last thing I would look at. If you said you cannot hear the difference when changing the bias, I believe something else is the bottle neck that you cannot even hear the improvement. The system sounds only as good as the weakest link in the system. My Acurus was biased quite cold, like 13mV across the emitter resistor. When I increase the bias, there is a distinct improvement. I am running 100mA per pair now. I did not change the resistor because I am lazy, and also at my listening level, I never get out of Class A, so even if I did not do the Oliver's optimization, I can get away with it for now.

JMHO.

Edit:
I look at the schematic. One thing stood out to me is it's a 120W amp, but with only 1 pair of output transistors. I am no expert, but I am not comfortable with this. I don't know what you can do. I would have used at least 3 pairs. May be I am way over kill, but at least I would like to see 2 pairs. Hell, I am OCD, I am using 5 pairs in my power amp and I am doing less than 100W but with 200mA per pair to give like 10W of class A power. In your case, you might want to consider MJW1302/3281 that has a bigger package and designed to handle more power.
Click to expand...

Alan ... i will take your words in considration just after i adjust my bias. thank you anyway
 
#23
Ian Finch said:
We are getting over-complicated here with irrelevant issues and operating conditions. This is a popular, proven DIY design here by member Apexaudio. The issue is with this particular build where the bias adjuster doesn't seem to work according to the voltmeter measurement but somehow the bias current must be varying because the heatsink temperature varies (assuming this is not a warm-up time delay matter).

So, either your meter is measuring incorrectly, the wrong channel in a stereo arrangement is being adjusted or there is a strange circuit error. You could be measuring at the appropriate test points but adjusting something that is unrelated. Check the soldering and terminals of P1 so that the correct terminals and connections according to the schematic are used.

Try this variation: Measure across both emitter resistors. That is, between the emitters of Q12 &13 instead of from one emitter to common. Ensure nothing is connected to the output terminal and loading it.
Now:
Do you measure twice the voltage drop at 24 mV now?
Does the measurement vary now with P1 adjustment?
Does the temperature of the heatsink still vary in warming and cooling?
Do you have Q10 attached properly to the heatsink or one output transistor case?
Click to expand...


i will try and double check again .. but i just installed a short link instead of the fuses ... do i have to remove them and replace with a 10R to measure the bias again or i just disconnect the speaker(load) and short the input and connect it to the star ground ......
 
#24
ostripper said:
Apex14 = baxandall based. While I have even seen a AX14 run perfect at 300W
(still4given), on one of my output stages , unless everything is perfectly sourced ....
this topology WILL oscillate if all is not "right on".
One offspec semi will "excite" the baxandall , I always recommend double checking
the BCxxx's .... no flux , and other extra precautions with this setup.
The driver and output stage choices also count heavily in this amps ultimate stability -
(super-pairs are different , especially as VAS).

OS.
Click to expand...

Ostripper ... iknow it's proven working and sure it's ok... thats why i was asking what is the mistake i'm making that makes the amp work but cant measure bias current and was checking the right procedjure to adjust the bias while i'm not a professional so i'm asking alot..but thank you anyway
 
#25
Well , that is the other possibility. Measure between output emitters , millivolt changes.
Measure between driver emitters , 1V to 1.5V should be the range (EF2).

If this does not happen , what I stated is a possibility. I stated that the AX-14 type
amp is a little more "touchy". My ONLY failure was with a super pair. 😱

Did you ask sir APEX ??

OS
 
#26
ostripper said:
Well , that is the other possibility. Measure between output emitters , millivolt changes.
Measure between driver emitters , 1V to 1.5V should be the range (EF2).

If this does not happen , what I stated is a possibility. I stated that the AX-14 type
amp is a little more "touchy". My ONLY failure was with a super pair. 😱

Did you ask sir APEX ??

OS
Click to expand...

i willd do measure the driver emitters voltage.... and yes i asked mr.Apex at ultimate 100w high fidelity thread ... and he is just ignor me ..
 
#28
VinDiesel said:
i will try and double check again .. but i just installed a short link instead of the fuses ... do i have to remove them and replace with a 10R to measure the bias again or i just disconnect the speaker(load) and short the input and connect it to the star ground ......
Click to expand...
Never begin testing and setup of amplifiers with a speaker or fixed load connected. You can surely place a 1-10k resistor across the input or short to signal ground but it may not be essential.

The 10R resistors protect the amplifier from high current damage at the testing stage (now) and provide a measuring point for total current. Unless your amplifier operates correctly first, you should not remove them. Wait until it functions correctly.
 
#29
i will try and double check again .. but i just installed a short link instead of the fuses ... do i have to remove them and replace with a 10R to measure the bias again or i just disconnect the speaker(load) and short the input and connect it to the star ground ......
Click to expand...

Can you place 10R resistor instead of fuse and post voltage across 10R resistor while adjusting bias.

Sonal.
 
#30
Test as here.
 

Attachments

  • APEX AX14 wireing test idle current.JPG
    APEX AX14 wireing test idle current.JPG
    305.2 KB · Views: 906
#32
Ian Finch said:
Never begin testing and setup of amplifiers with a speaker or fixed load connected. You can surely place a 1-10k resistor across the input or short to signal ground but it may not be essential.

The 10R resistors protect the amplifier from high current damage at the testing stage (now) and provide a measuring point for total current. Unless your amplifier operates correctly first, you should not remove them. Wait until it functions correctly.
Click to expand...

okay ian .. iwill install them a gain and try
 
#35
by the way Ian.. i measured voltage across each emitter resistor 12mv and when measured both together from lead to lead measured no thing at no input signal and no load.

i also measure the voltage at output speaker terminals it measure 12V...oooooooh sh..
(no load and no input signal)

what is going on???!!!!
 
#36
VinDiesel said:
.. i measured voltage across each emitter resistor 12mv and when measured both together from lead to lead measured no thing at no input signal and no load.

i also measure the voltage at output speaker terminals it measure....12V...
Click to expand...
OK. Now we begin to see problems you really have. It could be soldering, damaged transistors or a pinout error (some small transistors that are fake and even good, unbranded copies may have different pinouts which won't be known unless you test all parts before assembly) Some constructors had problems with fake 2N5401 pinouts. Test all transistors in-circuit by powering up and measuring B-E junction voltages which should be 0.6-0.7V (either polarity) if they are good.

Note that a wrong reading could be due to assembly error, failure, other circuit error, even a fake part (unlikely if you bought from a good source like RS) Your meter is quite good and accurate enough for any tests required but it lacks a transistor test feature. This is helpful for inexperienced constructors but you can test with the meter well enough or buy a cheap tester on Ebay that will likely work very well if you cannot easily build something from the >100 designs on the internet and some here too, in the test forum.

The thread you posted in - 100W Ultimate Fidelity Amplifier - has actually gone completely off-topic. People are posting now about many models like A9, A10, A14, A14X so maybe you should be specific which model you have problems with and post the results you have, like this 12V offset. It is much easier for Mile to deal with facts supported by data, even a schematic with voltages clearly marked, as others have posted.

First, I would re-read the first 100 posts or so, to get a clearer, simpler picture and answers like the first constructors had.
Test the voltage across the 10Rs and tell us what you have too.
 
#37
Hi Vin,

Sorry you didn't get help in the AX14 thread. I was busy trying to solve the A10. If you are not seeing a change in bias turning the trimmer then there is something wrong. I have built three of the AX14 as well as the SX14 and NX14 which all have the same front end. If you have a diode test on your meter, try testing each transistor to see if you have a bad one or solder bridge somewhere. If that shows all of them good then we are going to need to see some voltage readings.

Blessings, Terry
 
#38
Ah , seems like sir Apex has too many models , as well. 😀

That's why I'm reviewing the slew stuff , too many choices - needs to
be more "structure". And more documentation -
Something like the "Badger" is always the same - easy to advise on a problem.
With many designs , one might get "forgotten" 🙁 .

OS
 
#39
The AX-14, being the first in that thread, is well documented with many of them built and tested. It is not difficult to get working. The OP here has something wrong. Either a wrong part or orientation or a solder bridge or cold joint. If built per the docs it will work.
 
#40
still4given said:
Hi Vin,

Sorry you didn't get help in the AX14 thread. I was busy trying to solve the A10. If you are not seeing a change in bias turning the trimmer then there is something wrong. I have built three of the AX14 as well as the SX14 and NX14 which all have the same front end. If you have a diode test on your meter, try testing each transistor to see if you have a bad one or solder bridge somewhere. If that shows all of them good then we are going to need to see some voltage readings.

Blessings, Terry
Click to expand...

Thank you Terry ... i hope you finished the A10 well... i will check the driver and pre driver transistors with the meter and tell you.... i can understand from your words that you doubt these transistors .. i will check and tell you then tell me what readings you need.

Thanks alot ....
 
Status
Not open for further replies.
Top Bottom