Yes, much stronger. 2sa1302/2sc5200 (for output stage) are definitively not a substitute for 2sa1011/2sc2344 (for driver stage). 2sa1011/2sc2344 have only 25W. And one more thing, 2SA1302 is complementary with 2SC3281 and 2SC5200 is complementary with 2SA1943.
Thanks
Thank you. I saw that on 2sa1302's specs sheet. What was pacificsemi.com thinking? Funny thing, I now have a bunch of good 2sa968 from various places.
Bad Buy
Yes, I agree. AndrewT, I am learning this whole new field--purchasing thru electronic project building. I said earlier that I am a retired nut who is watching other retirees dying left and right because they don't know what to do with themselves. " If you are what you do while working, who are you when you are not working?" If I lose a few dollars while learning this whole enterprise of electronics, so be it. I just ask for your patience.
What is actually the motivation for substituting the MJ15003/4's with different devices? The MJ15003/4's have better thermal conductivity than TO264 devices which is pretty critical for this project. They are also almost everywhere available. Offcourse they have lower Ft, but I would consider that as a advantage, this will reduce chances of oscillation a bit more. Can someone explain which specific parameters are important and why?
I owned a KMA100 years ago 87-88, and if my memory still serve me well.. the output transistors had a uniqe Krell code, and unlike any motorola printing it had green-black ink on the PNP-NPN parts,which reminds me more of japanese part marking.
but then again, it may be only my particular units had these devices...
So i wouldn`t be so sure as to note that early Krell used MJ15003/4 parts.
I know for sure the 1995 models use the motorola MJ15024/5 .
but then again, it may be only my particular units had these devices...
So i wouldn`t be so sure as to note that early Krell used MJ15003/4 parts.
I know for sure the 1995 models use the motorola MJ15024/5 .
Hi BenY, Krell like many manufacturers had parts labeled specific for them so that one could not just buy a spare over the counter.
These parts where higher performance standard parts guaranteed a minim specification, so Dan may have specified gain of some number, Ft and so forth, so it was basically a OE part.
These parts where higher performance standard parts guaranteed a minim specification, so Dan may have specified gain of some number, Ft and so forth, so it was basically a OE part.
fakes
homemodder, I just got into this fantastic amp and will not anyone tear-it-down.
OK Sir!!!!! How do you go troublefree in these free-markets? You may have a place to purchase "real transistors," but good luck. See Kaplaars, I am learning to tell some else--which is, I believe, one word.
homemodder, I just got into this fantastic amp and will not anyone tear-it-down.
performance
would you lose performance using or making a device where you could slip and slide out TO-220s to sit make in your lazy chair a get a whiff of difference?
Hi Friends,
It's a while ago I posted some updates, very busy period. Although I still have vacation, I worked a bit to finance the KSA inter alia to other projects haha.
I've made some progress. I've also red quite a bit of literature to learn more about the stuff.
First I wanted to construct a Zobel network, but I calculated that using an 1 Watt resistor is going to be tight. Although there is some breating space, and 20 KHz at full power (most extreme circumstance for the Zobel when the amp does not oscillate) is not going to occur too often, I really want to make it safe, so I will order some non-inductive resistors which can withstand a bit more power (3W). Devices which can withstand more power are almost always inductive, which is a no-go for the Zobel. Offcourse, when it oscillates bad even the 3 W resistor will go down no matter what, but at least using a 3 W device will buy a bit more time before it goes down.
I've also obtained some nice connectors for the speaker terminals and 6mm thick wires:
And connected the speaker terminals parallel with each other and the speaker protection board (negative leads are not connected in the picture).
I've also bought some more quiet fans. They are absolutely inaudible. They are from Noise Blocker Germany. Got a few rubber rings with it so vibrations can't be transferred to the cabinet. Really nice stuff, but I am afraid airflow is reduced to much, so maybe I change them for fans that have a bit more airflow:
Andddddd I've been thinking about some way of fan controll with temperature readout. I wanted to keep it modular and I wanted to controll 4 fans with it because I want to place two extra additional 12 cm fans underneath the cabinet for extra airflow. I found a nice device called the Scythe Kaze master:
But I did not wanted to add the controller + PSU in the cabinet of the amplifier. I don't think chances of distortion caused by this device are big, but still it is another transformer and a few bits of electronics extra next to the amplifier. So I was thinking and thinking and finally tadaaa I bought an external DVD-RW drive. Not to make DVD's but I could use the cabinet and PSU. It has exactely the PSU I need, 2A @5VDC and 2A @12VDC:
Nice thing about the cabinet is that the Kazemaster fits perfectly in it! But it needed some tweaking; it had some scratches so I sprayed the cabinet over with some silver spray paint and I added some new feet underneath it:
Resulting in:
It is not finished yet. I have to construct a connection between the KSA and this device. I am thinking of using D-sub connectors. Have a lot of them which I collected a few years ago when I repaired and scrapped computers.
Loads of pictures, and a few updates, but it is not finished yet
@ Spurtle: I would say that you don't loose performance. But it can occur that the contacts of the device corrodes which makes the transision resistance a bit higher.
It's a while ago I posted some updates, very busy period. Although I still have vacation, I worked a bit to finance the KSA inter alia to other projects haha.
I've made some progress. I've also red quite a bit of literature to learn more about the stuff.
First I wanted to construct a Zobel network, but I calculated that using an 1 Watt resistor is going to be tight. Although there is some breating space, and 20 KHz at full power (most extreme circumstance for the Zobel when the amp does not oscillate) is not going to occur too often, I really want to make it safe, so I will order some non-inductive resistors which can withstand a bit more power (3W). Devices which can withstand more power are almost always inductive, which is a no-go for the Zobel. Offcourse, when it oscillates bad even the 3 W resistor will go down no matter what, but at least using a 3 W device will buy a bit more time before it goes down.
I've also obtained some nice connectors for the speaker terminals and 6mm thick wires:
An externally hosted image should be here but it was not working when we last tested it.
And connected the speaker terminals parallel with each other and the speaker protection board (negative leads are not connected in the picture).
An externally hosted image should be here but it was not working when we last tested it.
I've also bought some more quiet fans. They are absolutely inaudible. They are from Noise Blocker Germany. Got a few rubber rings with it so vibrations can't be transferred to the cabinet. Really nice stuff, but I am afraid airflow is reduced to much, so maybe I change them for fans that have a bit more airflow:
An externally hosted image should be here but it was not working when we last tested it.
Andddddd I've been thinking about some way of fan controll with temperature readout. I wanted to keep it modular and I wanted to controll 4 fans with it because I want to place two extra additional 12 cm fans underneath the cabinet for extra airflow. I found a nice device called the Scythe Kaze master:
An externally hosted image should be here but it was not working when we last tested it.
But I did not wanted to add the controller + PSU in the cabinet of the amplifier. I don't think chances of distortion caused by this device are big, but still it is another transformer and a few bits of electronics extra next to the amplifier. So I was thinking and thinking and finally tadaaa I bought an external DVD-RW drive. Not to make DVD's but I could use the cabinet and PSU. It has exactely the PSU I need, 2A @5VDC and 2A @12VDC:
An externally hosted image should be here but it was not working when we last tested it.
Nice thing about the cabinet is that the Kazemaster fits perfectly in it! But it needed some tweaking; it had some scratches so I sprayed the cabinet over with some silver spray paint and I added some new feet underneath it:
An externally hosted image should be here but it was not working when we last tested it.
Resulting in:
An externally hosted image should be here but it was not working when we last tested it.
It is not finished yet. I have to construct a connection between the KSA and this device. I am thinking of using D-sub connectors. Have a lot of them which I collected a few years ago when I repaired and scrapped computers.
Loads of pictures, and a few updates, but it is not finished yet
@ Spurtle: I would say that you don't loose performance. But it can occur that the contacts of the device corrodes which makes the transision resistance a bit higher.
Hi Friends!
How are the other builds progressing? Hope the other builds are progressing prosperous
I've made little progress, progress is going with baby steps right now, new year at university started. We started right away with quantum mechanics, which I have to say is somewhat difficult. But I am lucky, even Richard Feynman said no one could understand quantum mechanics, so it is not entirely my fault But back on topic Kaplaars!
I've added a D-sub connector to the external fan control unit. Therefore I had to cut a little bit out of the backplate which is attached to the cabinet. I also had to solder all the wires to a re-used D-sub connector which I scrapped from an old computer.
Mounted it to the backplate:
And the cabinet:
I am waiting for a thick cable with 25 small wires in it. I will use this cable to form a connection between the KSA chassis and the fan control unit. I also have to add the other D-sub chassis part to the KSA-100 chassis. So still work to do. But if everything turns out right it should be finished before Christmas
I hope
How are the other builds progressing? Hope the other builds are progressing prosperous
I've made little progress, progress is going with baby steps right now, new year at university started. We started right away with quantum mechanics, which I have to say is somewhat difficult. But I am lucky, even Richard Feynman said no one could understand quantum mechanics, so it is not entirely my fault
But back on topic Kaplaars! I've added a D-sub connector to the external fan control unit. Therefore I had to cut a little bit out of the backplate which is attached to the cabinet. I also had to solder all the wires to a re-used D-sub connector which I scrapped from an old computer.
An externally hosted image should be here but it was not working when we last tested it.
Mounted it to the backplate:
An externally hosted image should be here but it was not working when we last tested it.
And the cabinet:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
I am waiting for a thick cable with 25 small wires in it. I will use this cable to form a connection between the KSA chassis and the fan control unit. I also have to add the other D-sub chassis part to the KSA-100 chassis. So still work to do. But if everything turns out right it should be finished before Christmas
I hope
Hi Friends!
This amp is getting more and more heavy. Who needs dumbells when you got an amplifier like this So I thought some handles would be handy. I also felt like there was something missing at the rear of the amplifier. Some nice handles should not suit ill. 1 + 1 = 2, so I wanted to mount some handles at the rear of the amplifier. A very friendly radioamateur offered me two very nice handles for free! They had some scratches, but they surely had potential. They are really strong, thick, iron handles. So, I sanded and repainted them with silver spray paint. I am very statisfied with how they turned out:
Next step was to mount them to the rear of the cabinet. This is not easy since it is difficult to predict were I had to drill the holes. But after some measurements I got it right. So it was time to drill the holes. Bingo, they fitted like a glove. I think it turned out really nice. The amp is becoming more and more professional
Rear:
And with the backplate mounted to the cabinet:
I am thinking of mounting two more handles at the front too. But I first have to find two nice handles before I, offcourse, can mount them
This amp is getting more and more heavy. Who needs dumbells when you got an amplifier like this
So I thought some handles would be handy. I also felt like there was something missing at the rear of the amplifier. Some nice handles should not suit ill. 1 + 1 = 2, so I wanted to mount some handles at the rear of the amplifier. A very friendly radioamateur offered me two very nice handles for free! They had some scratches, but they surely had potential. They are really strong, thick, iron handles. So, I sanded and repainted them with silver spray paint. I am very statisfied with how they turned out:An externally hosted image should be here but it was not working when we last tested it.
Next step was to mount them to the rear of the cabinet. This is not easy since it is difficult to predict were I had to drill the holes. But after some measurements I got it right. So it was time to drill the holes. Bingo, they fitted like a glove. I think it turned out really nice. The amp is becoming more and more professional
Rear:
An externally hosted image should be here but it was not working when we last tested it.
And with the backplate mounted to the cabinet:
An externally hosted image should be here but it was not working when we last tested it.
I am thinking of mounting two more handles at the front too. But I first have to find two nice handles before I, offcourse, can mount them
Wow, it is quiet here, time to shake this treat up
Another update concerning the KSA project, I've finally connected the Zobel network. I received some parts a few days back; two nice 4W MOX 5.6 ohm resistors and two nice 100 nF Audyn caps. I haven't got a clue how the Audyns perform. But hey were not expensive and easy to mount, so I ordered them with the resistors.
Nicest way to mount them would be on a little PCB, but that would made the leads longer, and since the leads of the Zobel need to be as short as possible, I mounted it directly to the speaker terminals:
But the Zobel raised a question: some power amplifier designs don't use a Zobel network, but only mount a capacitor parallel to the speaker terminals. But is this not dangerous? I am asking this because, when an amplifier oscillates at an high frequency, the capacitor gets more and more conductive, thus I would say eventually when the frequency is high enough it can act like a short circuit between the leads. This would be deadly for any amplifier.... well that is what I think. Is there something I am missing in here? Does this work because the speaker itself is still parallel to the Zobel which adds some resistance? And that impedance changes with frequency, so the higher you go in frequency, the more impedance the speaker will have because of it's coil? I would say no, because current takes the shortest lead, which is at high frequency, through the capacitor. Also it would be still dangerous when there is not a speaker connected to the amplifier. Can someone help me out on this one?
Another update concerning the KSA project, I've finally connected the Zobel network. I received some parts a few days back; two nice 4W MOX 5.6 ohm resistors and two nice 100 nF Audyn caps. I haven't got a clue how the Audyns perform. But hey were not expensive and easy to mount, so I ordered them with the resistors.
An externally hosted image should be here but it was not working when we last tested it.
Nicest way to mount them would be on a little PCB, but that would made the leads longer, and since the leads of the Zobel need to be as short as possible, I mounted it directly to the speaker terminals:
An externally hosted image should be here but it was not working when we last tested it.
But the Zobel raised a question: some power amplifier designs don't use a Zobel network, but only mount a capacitor parallel to the speaker terminals. But is this not dangerous? I am asking this because, when an amplifier oscillates at an high frequency, the capacitor gets more and more conductive, thus I would say eventually when the frequency is high enough it can act like a short circuit between the leads. This would be deadly for any amplifier.... well that is what I think. Is there something I am missing in here? Does this work because the speaker itself is still parallel to the Zobel which adds some resistance? And that impedance changes with frequency, so the higher you go in frequency, the more impedance the speaker will have because of it's coil? I would say no, because current takes the shortest lead, which is at high frequency, through the capacitor. Also it would be still dangerous when there is not a speaker connected to the amplifier. Can someone help me out on this one
?
Last edited:
the Thiele Network that is added to give the amplifier enhanced HF stability comes in two forms. There is also an infinite range between the two forms.
1. R+C at the amplifier output, then L to the speaker terminal.
2. L//R to the speaker terminals, then C across the speaker terminals
Dr Cherry showed how to calculate the component values for the two Thiele Networks and also how to calculate for the whole range between the two limiting values mentioned above.
Do not simply put a cap across the speaker terminals. That alone will make stability worse, not better.
You must put in the whole Thiele Network.
However, some designers can make their amplifiers stable without needing the Thiele Network. Those designers are clever and don't need DIYaudio to teach them how to stabilise amplifiers.
It's us that need all the assistance that Members provide. Use a full Thiele Network, until YOU can design out some or all of the Thiele components.
1. R+C at the amplifier output, then L to the speaker terminal.
2. L//R to the speaker terminals, then C across the speaker terminals
Dr Cherry showed how to calculate the component values for the two Thiele Networks and also how to calculate for the whole range between the two limiting values mentioned above.
Do not simply put a cap across the speaker terminals. That alone will make stability worse, not better.
You must put in the whole Thiele Network.
However, some designers can make their amplifiers stable without needing the Thiele Network. Those designers are clever and don't need DIYaudio to teach them how to stabilise amplifiers.
It's us that need all the assistance that Members provide. Use a full Thiele Network, until YOU can design out some or all of the Thiele components.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.