Krell KAV-250 tuneup

[rdnzl999]

My new (old) KAV-250 arrived today, now driving my old (were new) reQuests.

I popped the top, all looked to be in order, none of the smoke appeared to have already escaped. It has a thin layer of British Columbian dust on it, seems never to have been opened, no paw prints inside.

Moved it into position in the pseudo-audio rack, powered it up, still no smoke. Measured 77w idle (cold), and 0.2 dc offset on the positive output of the right channel.

Power off, hooked up the speakers and RCA inputs, powered it on, and heard an annoying hum in both channels.

Hmmm. Hum. Hmmm. Oh yeah, forgot to short the dangling pin 3 of the XLR inputs to ground. Used a bent paperclip inside one of my new (new) Neutriks (with which I will build a new interconnect). That took care of the hum. Hmmm.

Played some tunes, sounds good. It replaces a similarly aged Acurus A250, which purportedly has some Dan D'Agostino genes in it itself.

Warmed, it draws about 85w at idle (after running with music for two hours). The manual says 210. 95-100w with signal at moderate daytime casual listening levels.

Cold, it had 0.2v dc offset on the negative leg of the right channel (measured to ground). Now that it is warmed up (and the XLR pin 3 is shorted to ground) I see only 0.08v offset max. Don't know yet if the open XLR was causing the DC offset, doubt it.

The heatsinks are warm, both about 102 degrees (F) running in daytime casual listening mode.

Sounds good, waiting for the late night to hear the rest of the magic.

---

Question:

I haven't stared at it real hard, and have no schematic, and no other experience with Krell circuits.

I see two little blue pots in each channel, at R22 and R68 on the "outer" board. There is a tiny screw in the top of each one.

Anyone know the circuits? Got any other hot tips for this old device?

Here's a picture of the board...

http://screencast.com/t/PZPdhkd8AN4C



PS: It has had the fuse changeout per the recall -- bottom -- http://www.krellonline.com/support.html

.

 

[Tosh]

Did you say "smoke"?
Here's a thread where one blew up (twice) and another was in a fire!

I would dust yours off a bit better....

 

[rdnzl999]

Did you say "smoke"?
Here's a thread where one blew up (twice) and another was in a fire!

I would dust yours off a bit better....

Thanks for the link, I think I found what I was looking for there:

"Does the 250a have an adjustment to reduce output offset?

Yes, there are two trimmers in the amp, the one near MPQ6600A1 (Quad complementary pair) is for adjust the offset.

Another one near the bias transistor 15031 mount on the heat sink is for adjust bias. "

And, yes, a "recall" was issued in 2009 after like 9 years: Replace the 20amp output fuses with 8amp (done). I knew that before purchase.

Translation: Don't short the outputs.

 

[Tosh]

If you're going to play with any trimmers, you should practice on the Acurus A250 (until you know what you're doing). If it idles stone cold like my 100X3 did, there's room to raise the bias, which gives you more class A and sounds sweeter. I raised the bias from about 20mA (factory setting) to 360mA (4W of class A!) after I made new heasinks and PS (and other things), and it sounds a LOT better.

BTW, don't worry about any offset on that Krell when cold.

 

[rdnzl999]

Yes, the Acurus has a single trimmer. Did adjustment years ago, it idled unevenly, one sink a little hotter than the other.

Twist the dial and idle temperature would go from stone cold to blazing hot. I set both sides evenly (per voltage on the base of the output transistors) and idle heat -- about 15 degrees above ambient. I didn't have a watt meter back then.

I don't remember seeing (using an oscilloscope) nor hearing any definitive difference with the changes in the bias settings.

Now, measuring with a Kill-A-Watt meter --nice little tool for about $18 --it draws 30w at idle.

The Krell owner reference says 210w at idle, which bothered me a bit (it has no standby) and I tend to leave it on all the time because I get to use it all the time.

210w idle would cost $16/mo in electric. But, it measures a much more reasonable 84w (warmed) and still under 100w during typical listening, so the fee won't be so much.

At the moment, 70 hours runtime and 5.6kwh consumed - about $0.60... Or $6.30/month, some "critical listening", some TV, some background music while puttering around.

The offset stabilized after warmup to 4.6mv and 1.1mv, so that's good enough, using a Fluke 87 --on 4 1/2 digits setting.

It's an old amp, just bought second hand off eBay, but looks fine and sounds fine too.

I wiggled all the connectors I couldn't pry off, applied NO-OX ID-A to all that I could (as well as to the fuses), buttoned it up and let it play.

The Acurus aquitted itself very well, I was quite happy with it, and don't plan to sell it. It will go in the bedroom with the SL3. I was having a hard time imagining how the new amp would surpass it.

The Krell is just a bit better -- soundstage a little further removed from the speakers, a little more definition in the bottom. Everything subtly more solid. More scary sounds way off the speaker axis at times. Higher frequencies, I dunno, I have genetic HF loss.



I'll probably do this next time I feel ambitious:

1. Pop the fuses to both channel power inputs. Measure the idle draw of the power section. Should be real close to 0 after inrush.

2. Power the right channel, measure idle draw. Power the left channel, measure the draw. Power both and remeasure.

3. Eat a hotdog and look at the numbers.

4. Decide if they are out of balance and fix that. Heat sinks are the same temp now, so doubt it to need attention.

5. Guestimate the "Class A" range based upon the current draw.

Need to find some calculation for that estimation. Got links?

 

[Tosh]

from NP's paper Leaving Class A:
"Push-pull amplifiers generally operate in Class A mode up to a point where the output
current is twice the value of the bias current."

So in my case of 0.36A bias:
Class A point: 2(0.36) = 0.72A

Power = I^2 * R
Class A power into 8 Ohms = (0.72)^2 * 8 = 4W

 

[rdnzl999]

That quote led me to http://www.enjoythemusic.com/magazine/manufacture/0808/

Ok... That's a nice little read.

When I read your post I thought "bias current" meant current at the base of a bi-polar, not the flow from the rails through emitter and collector. I haven't studied circuits for a long time.

Thanks for hint leading to the link.