Krell KSA 100mkII Clone - Page 4 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 22nd April 2006, 08:29 AM   #31
jam is offline jam  United States
diyAudio Member
 
jam's Avatar
 
Join Date: May 2001
Location: Auburn, CA, USA
The pre 1983 KSA-100 is almost an idental copy of the Aragon amplifiers( Krell has a hand in design) except for the bias that the unit was set at,
The Aragon had an excellent layout, with everything on one PCB that was attached to the hesstsink.............if someone has a picture,it might be a good starting point.

Jam
  Reply With Quote
Old 22nd April 2006, 08:42 AM   #32
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Hi,
looking at the pre83 and Mk2 versions of the KSA100, it appears that they adopted ccs (just like AL) for the LTP and simply added 2pairs of output devices giving a higher voltage version of a KSA50. As someone pointed out a while back this sometimes reduces sound quality.

With the higher driver voltage the driver becomes the limit and this could be what causes returns under warranty and possibly sound quality.

Along comes Mk2 with the return of the Zener fixed LTP source and cascode for the VAS and doubling up the driver ability. But the big change is the intermediate dual complementary LTP FET stage.

The big question:- which of these changes made 100Mk2 better than KSA50 and which made it worse.

Can I suggest that we incorporate ALL the tweeks that Dan had, that make it better, and leave traces in the board for any that cannot be shown to improve sound quality thus making these optional (the PCB space for this will be marginal) and allowing reversion to KSA50 topology where builders can experiment to find the best sound.

Looking back and comparing my driver posting, Dan has 1driver (MJE1503x) to 500W of output devices in the KSA50Mk2 and KSA100Mk2 cf. my limit of: 1driver per 750W of output device and Dan's 100pre83: 1driver per 1000W

The other difference, the driver Re has gone from 25r in KSA50 to 47r in 100pre83 to two times 75r in 100Mk2.
__________________
regards Andrew T.
  Reply With Quote
Old 22nd April 2006, 09:08 AM   #33
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Default Vbe temp compensation

Hi,
this has been discussed before and I have deliberately NOT looked back at what was posted.

My thoughts are, the output stage is heavily biased and the running temperature depends little on the output power and much more on the ambient temperature. Once up and running and fully warmed up the output stage temperature and the consequent Vbe of the output transistors will vary very little and should not need much in the way of temperature compensation. Even when the Iq varies a little due to ambient changes with the seasons the effect on the total ClassA output will be small and consequently the effect on sound quality will be even smaller.

The driver stage is initially in ClassA and it too will depend on ambient temperature. But, the big differenece here is that the drivers are physically smaller (respond more quickly) and their temperature (Tj) depends much more on the output current. The variation in Vbe of the drivers is likely to be many times higher than the variation of the output stage Vbe (whether it is two times or ten times I cannot analyse nor measure). The Vbe multiplier if closely connected to the drivers Tj can and should compensate for the driver Vbe variation. There is little need to artificially raise the driver Tc to match the output stage heatsink temp and reduce the drivers' ability to cope with it's own loading.
I can see a possible need to optimise the driver Tc and Tj but this could only be done if it were separated from the output heatsink and then driver Re could be used to experiment with optimum level of driver Iq (controlling Tj + Tc).

If KSA100Mk2 is actually a high bias ClassAB design then a small amount of overcompensation in the driver stage will reduce output Iq as power levels rise above the ClassA level. This effect will be quite slow since the output stage has the inertia of eight Tc and a large heatsink to overcome before temperatures start to react. But at least overcompensation at high driver outputs will be in the correct direction if a little output compensation is required.

So my suggestion is that the four drivers and Vbe multiplier be on a common heatsink that strives to keep driver Tc considerably below output Tc.
__________________
regards Andrew T.
  Reply With Quote
Old 22nd April 2006, 09:09 AM   #34
awpagan is offline awpagan  Australia
diyAudio Member
 
Join Date: Nov 2004
Location: sydney
Quote:
Originally posted by AndrewT
Hi,
looking at the pre83 and Mk2 versions of the KSA100, it appears that they adopted ccs (just like AL) for the LTP and simply added 2pairs of output devices giving a higher voltage version of a KSA50. As someone pointed out a while back this sometimes reduces sound quality.

With the higher driver voltage the driver becomes the limit and this could be what causes returns under warranty and possibly sound quality.

Along comes Mk2 with the return of the Zener fixed LTP source and cascode for the VAS and doubling up the driver ability. But the big change is the intermediate dual complementary LTP FET stage.

The big question:- which of these changes made 100Mk2 better than KSA50 and which made it worse.

Can I suggest that we incorporate ALL the tweeks that Dan had, that make it better, and leave traces in the board for any that cannot be shown to improve sound quality thus making these optional (the PCB space for this will be marginal) and allowing reversion to KSA50 topology where builders can experiment to find the best sound.

Looking back and comparing my driver posting, Dan has 1driver (MJE1503x) to 500W of output devices in the KSA50Mk2 and KSA100Mk2 cf. my limit of: 1driver per 750W of output device and Dan's 100pre83: 1driver per 1000W

The other difference, the driver Re has gone from 25r in KSA50 to 47r in 100pre83 to two times 75r in 100Mk2.

the big question...................
Was the ksa100 better than the ksa50?

It had more power so it could drive low impedance speakers (apogee) but what about the sound?


Andrew
i agree with the possibilty of leaving things out.

everyones ears are different.

allan
__________________
Indecision makes the world gone round.
Maybe
  Reply With Quote
Old 22nd April 2006, 09:35 AM   #35
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Hi Mark,
do you have access to Krell specifications?

160W into 8r needs an output of 50v6pk. This is impossible on a dual polarity Vrail= 50Vdc and even less possible if the +-Vrail=48Vdc when fully biased.

A maximum of 120W into 8r is more likely for +-48Vdc and possibly 130W for +-50Vdc. These maximum powers would be at clipping level when distortion is likely to be quite high.

What are the bias settings for the 100pre83 and 100Mk2?
Is there a later (Mk3) release? How do the KMA numbers compare?

I (we) could do with a full set of comparative data before we second guess how the Klone should be designed. It's like selecting the timber to build a bridge but then discover we haven't measured the width of the river.

Awpagan,
a little bit of ingenuity can keep the options open, a bit like Jan's Klone with the VI limiting (use it if you feel the need) or All's piggy back add ons for the very high voltage 50Klone.
__________________
regards Andrew T.
  Reply With Quote
Old 22nd April 2006, 10:19 AM   #36
diyAudio Moderator
 
pinkmouse's Avatar
 
Join Date: Apr 2002
Location: Chatham, England
Exactly. I don't have the time to do so much development on this one, so if you guys can decide what you want in advance, and get the schematic and devices nailed down, then that makes life so much easier, and you might actually get boards this year!
__________________
Rick: Oh Cliff / Sometimes it must be difficult not to feel as if / You really are a cliff / when fascists keep trying to push you over it! / Are they the lemmings / Or are you, Cliff? / Or are you Cliff?
  Reply With Quote
Old 22nd April 2006, 12:01 PM   #37
diyAudio Member
 
Join Date: Aug 2003
Location: Prague,Czech Republic
Can you tell me, guys, for what do you still can a Krell's ? Personally all Krell's, which I have heard, had audible distortion... Or is main advantage, that they can drive one ohm speaker ? Are you building amp or velding machine ? BTW, one ohm speaker can design only moron....
  Reply With Quote
Old 22nd April 2006, 02:25 PM   #38
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Hi Upupa,
if your
Quote:
ohm speaker can design only moron
can do this, then I'll buy ten of them.

Sorry, cheap shot you don't deserve. It's the in me that could not desist.
__________________
regards Andrew T.
  Reply With Quote
Old 22nd April 2006, 02:47 PM   #39
diyAudio Member
 
Join Date: Sep 2001
Location: Salt Lake City
Quote:
Personally all Krell's, which I have heard, had audible distortion...
Actually the open minded view that I've always taken is to say that ALL ampifiers have audible distortion of some sort. The distortion of a giiven amplifier is a certain color and it just depends on what color you like. There is no perfect amplifier . If we could just get Upupa Epops to contribute to the thread in a positive manner......

Andrew,
I agree with your take on the 100's power output. It couldn't be more than 120 watts. I don't have any info on the KMA100... sorry. I doubt that it had any more power output than the stero version which was basically dual mono. I have a picture of the inside of a KMA-100 which I will look for and post later today.

As for the driver and VBE sensor I agree. I just found mounting them on the main sinks alot easier. Also the JAN board had the drvices tabs reversed from each other and the VBE in another position away from the drovers making even mounting to a small common sink a tough job, especially for newbies. Keeping in mind that most of us have fan cooled our amps and mounted the drivers and VBE on the main sink temperature changes and tracking can still occur pretty rapidly but granted not as fast as the smaller tab devices can. Now convection cooling would be another matter and it would happen very slowly!

I am presently re-building my 50 with some of the original heat sink tunnel that Krell used in the production models so my drivers and VBE will be mounted the way you originally pointed out.

Everyone,

As for the PCB if we make up a MK-2 board straight over it would be easily reverse-adaptable to building any of the earlier KSA-100 circuits. Neither myself, Al, or my friend Dave have time to adapt or figure out a universal pcb to accomodate all the versions. I am only willing to foot the bill to have 100 boards made up. Its easy to envision that the Mk-2 board straight over can be very adaptable in itself just by jumpering out certain devices and changing some part values to building the earlier versions of the amp. Hopefully those that are going that route would take the time to share that in a future chart with all of us. However it should be up to each individual building the amp to figure out for him self..... Otherwise this PCB deisgn might go on for two years! So in simple terms the universal board would already exist in the from of the MK-2 board!!

As for parts the only obsolete items are VN0210N5 and VP0210N5 devices. These drvices are only drwaing 3 mils so they an be replaced with Zetex ZVP3310 and ZVN3310 as JWB did in his KMA 160 reverse engineered Krell. HE did not even bother to sink them together since they dissipate so little. So that being said all that need be changed are those devices over to the Zetex unless anyone knows of a better device (I had originally thought of IRF9610 and 610 but the Zetex's spec better). All other parts are readily available off the shelf.


Al,

So that being said how do you feel about doing the Mk-2 board straight over, perhaps adding 2 spots for extra bypass on the main board like you did for the 50 main board. Then just expand on the present KSA-50 driver board? The O.P. boards shouldn't be any big deal and could come last.


Mark
__________________
KSA-100 WIKI
  Reply With Quote
Old 22nd April 2006, 02:54 PM   #40
diyAudio Member
 
jacco vermeulen's Avatar
 
Join Date: Oct 2004
Location: At the sea front, Rotterdam or Curaçao
Send a message via Yahoo to jacco vermeulen
I believe the MKI had 41Vac secondaries, which is 57 Vdc something after the bridge.
The MKI powersupply states 52Vdc at the rails, which should stand for the loaded rail voltage.
160 watts continuous at 1% distortion in 8 Ohms should be in reach of these 52Vdc rails.
The MKII is unlikely to have lower rail voltages than the MKI.

I still have the 1986 test article of the KSA100MKII, the same Loek posted some time ago.
KSA100; 160 in 8 Ohms, 280 in 4, 400 in 2.
The KSA100 was measured to have a soft clipping behavior between 145 and 160 watts in 8 Ohms.

btw: i've heard both the kSA50 and the KSA100 on proper loudspeakers.
The KSA100MKII may have been better on Apogee's, i've never understood the charm of those metal washboards.
__________________
The buck stops Here
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 08:25 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2