• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

200WPC MOSFET Kenwood KR-9600

EchoWars

Hiding in Honduras
Sponsor
Any interest in watching a dead KR-9600 revived as a 200WPC MOSFET receiver? :yes: :yes:

Amplifierguru (Greg Ball) has developed a rather nice MOSFET amp topology, christened the SKA, or Simple Killer Amp, using IRFP240/IRPF9240 International Rectifier HEXFET's. He has two basic amp kits available, the GB150D, using two pairs of output devices per channel and supply voltages of about +/-55V for outputs of 150W or so, and the GB300D, using four pairs of outputs per channel and capable of 300W with supply voltages of about +/-75V or thereabouts. Feedback from those who have built the amp is very positive.

I have had SPL db's dead KR-9600 here for a while, hoping to either luck out and run across some good output modules, or find a way to rebuild the output stage either using a small fabbed circuit board and discrete output devices, or perhaps fabricate a board and use an STK output. Any of these solutions would require much time and labor on my part. Yes, I know ilimzn and I came up with a replacement for the TA-200W, but there isn't much room inside to do the work, and I can see this eating up a very large about of time to make work.

After Amplifierguru joined AK recently (have known him for a while from DIYAudio.com), I paid a visit to his webpage. What caught my eye was the very small size of the GB150D amp module...it appeared that there might be enough room to mount it on the stock KR-9600 amp heatsink. The question was whether it could operate at the +/- 67V rails of the KR-9600, so I shot off an email. A disappointing reply from Greg followed:
Sadly NO - the GB150D is on a stretch at +/-58V due to only 2 pairs 150W MOSFETs. Some would do it, but I like the amps to drive all phase angles within the SOA. Also the BC546C/56C input comp pairs would be right on their 65Vceo. The larger GB300D uses double the output MOSFETs (4 pair) and cascoded inputs for up to +/-75V and 300W. But it's 180mm long x 50mm. What would be perfect is one in between the two - which I don't make! Bugger.
Well, me being me, I wasn't ready to fold yet.

The GB300D has a board length of about 180mm. There's no way to adapt this to the stock 9600 heatsink, but looking at the ones available from Conrad Heatsinks, they have some that are 200mm long, and the stock 9600 heatsink just happens to be 8" long. I see no reason why I can't use some of Conrad's heatsinks to replace the stock 9600 one. Yes, it increases the price (the configuration of heatsinks I will build will run about $80), but in the end it is a tradeoff of either many hours of my work to make a discrete replacement, or buy the parts I need for a reasonable price and cut the labor needed to bring the 9600 back to life down to something reasonable.

Once the parts arrive, I plan on documenting the revival of this receiver. Follow along with me if you're curious, but don't expect the first installments for a bit...both the boards from Greg and the heatsinks will come from Australia, and there are two choices for shipping - fair priced and slow as molasses (probably two months), and 'bend over and grab your ankles' but here in a week. I'm afraid that the slow method is the one I'm going to have to use...

In the meantime anyone considering an amp project should check out Greg's page (linked above), and perhaps some of the threads - either at http://www.diyhifi.org or http://www.diyaudio.com . Greg himself is quite the personable soul, so if you have questions or ideas or are just curious, send him a PM or email.

I'll try to be back soon with updates.
 
Register to hide this ad
One thing I failed to mention is the obvious...if it blows up with the International Rectifier HEXFET's in it, it can be fixed. Serious advantage compared to the stock 9600.

If my own 9600 bit the dust, this is the route I'd go without hesitation.
 
Sounds interesting

Hey Glenn sounds interesting :yes: . hmmmmm! :scratch2:
I have an extra 9600 sitting on the bench. I still haven't finished the one I'm working on yet though.

:lurk: I was looking at these guys a while ago but never found anyone that tried these. http://www.ampslab.com/hx200.htm

John
 
Last edited:
Those are OK modules, but I would not try to get, say, 200W out of four of those devices (he uses the same HEXFET outputs as Greg, and his comments above about the voltage rails apply regardless of the amp topology). Asking a little much. Price is a little whacked too. Greg's GB150D is comparable (in a way) to the Hx200, and is $100 cheaper. But neither one of those is going to be trustworthy at +/-67V, which is why I'm replacing the heatsink and using Greg's GB300D. It should be quite happy at the voltages supplied by the KR-9600 supply.

http://www.diyaudio.com/forums/showthread.php?s=&threadid=77073
 
keep in mind that every converted KR9600 puts one good STK module into the market. This trade offsets the conversion's cost. I, too have a 9600, and will watch!
 
Nakdoc said:
keep in mind that every converted KR9600 puts one good STK module into the market. This trade offsets the conversion's cost. I, too have a 9600, and will watch!
Not if the STK is one of the reasons why it is bad in the first place. Fun to watch.
 
Well, the heatsinks are ordered ($78.47), but my PayPal account was too depleted to get the amp boards on the way, so that'll have to wait till a bank transfer takes place.

The four heatsinks consist of two MF20-100's (200mm long x 100mm high), and two MF20-50's (200mm long x 50mm high). The 100mm heatsink will be the one that the devices mount to, and the 50mm heatsink will mount 'flat to flat' on the backside of the 100mm heatsink, above the mounted MOSFET's for additional cooling area. It'll have to be drilled and tapped, and smeared with heatsink compound, but hopefully the two of them together will be sufficient to do the job.
 
Possibly the single coolest project I have seen on here. Can't wait to see how it works. I am thinking you will have to bring it to the fest next year too. Who knows, the parts might have arrived by then! :)
 
Last edited:
Nakdoc said:
keep in mind that every converted KR9600 puts one good STK module into the market. This trade offsets the conversion's cost. I, too have a 9600, and will watch!
Yamaha B-2 said:
Not if the STK is one of the reasons why it is bad in the first place. Fun to watch.
Well, I got plenty of things happening here on a daily basis, and taking a working 9600 and doing a conversion to MOSFET would be pretty low on my list of 'things to do'. Not that it isn't a terrible idea, but rather it would be more of a project for someone with a 9600, a lot of time, and the technical skills to make it happen. In a case like this you'd certainly end up with a couple of good spare modules, but realistically this makes sense for replacing dead modules only. For me anyway...

I also have a 9600, and if I like the way this turns out, I may buy the parts and keep 'em in the wings as a hedge against the day that one of the output packs dies...

What I would really have liked to do would have been to use of of Conrad's MF20-151.5 heatsinks (200mm x 151.5mm) and have the top machined down to about 125mm, and 'back' it with a MF20-75 (200mm x 75mm). This would give a lot more cooling area, and might make it even made it easier to mount in the 9600. But the additional cost of the machining and having it powder coated again would push this particular project over the top in terms of feasibility. Then again, perhaps the best thing to do would be to talk to Conrad and see if you couldn't get them to cut you one the proper size in the first place.
 
EchoWars said:
Well, I got plenty of things happening here on a daily basis.
Yeah....really. :D

EchoWars said:
But the additional cost of the machining and having it powder coated again would push this particular project over the top in terms of feasibility. Then again, perhaps the best thing to do would be to talk to Conrad and see if you couldn't get them to cut you one the proper size in the first place.
Am wondering if you meant anodized. Powder coating would be putting a fairly thick polymer layer on the heatsink resulting in a significant reduction in its ability to get rid of the heat. Also, with today's good saws, it would be considerably quicker and easier to saw off the unwanted portion and then take one pass with a fly-cutter to smooth out. Perhaps you could get what you want pre-anodizing, saw/flycut and then have anodized.
 
The diference between anodizing and powder coating is pretty minimal (and anodizing is slightly better, yes)...the stock heatsinks from Conrad are powder coated.
 
Glenn,

I know you've already ordered the heatsinks, but was wondering if Technut could machine the
heatsinks that you would really like in there.

And since he's here in my hometown, which is only spiting distance from you as you know, the
cost might be less as well.

But I'll trust whatever decision you make. :thmbsp:

Scott
 
The ones mentioned above have been ordered and paid for. I really don't think that the additional heatsink area is totally necessary...I just would have liked to fill every bit of available space with heatsink. :D I also wanted to show that this will work fine with 'off-the-shelf' parts.

In all honesty, it didn't occur to me to ask Conrad if they'd custom-make a 120mm version for me, but I can guarantee that the costs would have gone way up (perhaps to $130 or so). Ordering the larger HS and having it machined was also a possibility, but again, I wanted to follow the KISS rule.
 
Do you suppose this modification will help or hinder the regulator issue?

At least that's the issue that I remember off the top of my head... the issue where the unit
runs pretty warm and you've had to place a heatsink where there wasn't one before. :scratch2:

Scott
 
Back
Top Bottom