Kenwood KA-801 bias and preamp DC Offset questions

Stack

Active Member
Hi,

I've recently recapped my KA-801 and wanted to fine tune things. I've had specific problems with the bias and DC offset, and was wondering if anybody could lend ke some friendly advice.

For the preamp DC offset (which I'm now thinking I should've left alone in the first place) the spec wants you to descend the lower rear panel, hook a DC meter to the test points, and turn the tiny screw until you get 0V.

My first problem is that when I hook the meter up and power on the amp (from a power strip without touching anything of course) the voltage reading fluctuates wildly, by as much as .75 volts! For about ten minutes it varies up and down until it settles in a fluctuation pattern of 30ish millivolts. Is this normal behavior? It makes it hard to get an accurate reading.

Second, the pots seem to be incredibly sensitive. Putting any force into a turn will move you by half a volt! I was able to get a sort of fine tune by shimmying the (insulated) eyeglass screwdriver with the tiniest inclination in the direction that I wanted, but is there a better way? Does it make sense to replace these pots with something a little less finnicky?

Finally, this might be just my imagination, but adjusting one channel seemed to mess up the other. After I thought I'd gotten everything square, I put the amp back together and the right channel (the first I adjusted) was off to the point of clipping! Back to the bench it went. In your experience, does it make sense to get a second meter so that both can be measured at once, or was this maybe user error?

For the bias, my problem is much more straightforward: the pots don't adjust down far enough to reach the 18mv reading that the maintenance manual calls for. These are 500 ohm pots. In this case, does it make sense to replace these with higher resistance (perhaps 1000 ohm) pots so they can be adjusted down? I'm not in a position to diagnose and replace all the resistors, transistors, and diodes in this amp right now (I just finished a full recap and don't have the patience at the moment).

The left channel can be brought down to 33ish mv and the right channel 22ish mv. What long term effects would happen with the bias this high? Does it make sense to "match" the channels up to 33mv or just keep them set to minimum?

Any help would be greatly appreciated! Thank you for your time.
 
...My first problem is that when I hook the meter up and power on the amp (from a power strip without touching anything of course) the voltage reading fluctuates wildly, by as much as .75 volts! For about ten minutes it varies up and down until it settles in a fluctuation pattern of 30ish millivolts. Is this normal behavior? It makes it hard to get an accurate reading.

We'll have to presume that you are performing these adjustments with no input signal connected, and at zero output volume, which is the correct procedure. It's also best to set the unit in AUX or TAPE mode so that the highly sensitive PHONO section is not feeding the preamp, potentially with background or "floor" noise. .

The balance of this differential amp circuit is also going to be sensitive to the input DC voltage. Measuring the 12V supply at Q3/Q5 might reveal an unstable power supply, which would make adjustment very difficult - About as difficult as walking through Cap Square without seeing a rainbow flag. If the supply is varying on startup and then settles down, the supply itself and its connections might need some attention. A little drift is somewhat acceptable, but jumping around is not.


Second, the pots seem to be incredibly sensitive. Putting any force into a turn will move you by half a volt! I was able to get a sort of fine tune by shimmying the (insulated) eyeglass screwdriver with the tiniest inclination in the direction that I wanted, but is there a better way? Does it make sense to replace these pots with something a little less finnicky?

They are going to be fairly sensitive in any case, but it may help to document the positions of the two trimmers, then clean (De-Oxit) and exercise them, lube them lightly, and return them to their approximate original positions. A multi-turn trimmer might make this easier, but a clean trimmer is certainly better than an oxidized/sticky one.


Finally, this might be just my imagination, but adjusting one channel seemed to mess up the other. After I thought I'd gotten everything square, I put the amp back together and the right channel (the first I adjusted) was off to the point of clipping! Back to the bench it went. In your experience, does it make sense to get a second meter so that both can be measured at once, or was this maybe user error?

Again, if the power supplies are affecting the adjustment, this might be a result. The same could be happening if those ultra-sensitive dual FET comparators are getting flaky.

A second meter might make the procedure easier, but each channel is adjusted one at a time regardless. However, that second meter might speed up the process and help reveal what happens to the opposing channel when one is adjusted. If you get a second meter, be sure it has a true low-millivolt scale, and it certainly would be best to get something with a pair of mini-grabber leads to keep things all connected simultaneously.


For the bias, my problem is much more straightforward: the pots don't adjust down far enough to reach the 18mv reading that the maintenance manual calls for. These are 500 ohm pots. In this case, does it make sense to replace these with higher resistance (perhaps 1000 ohm) pots so they can be adjusted down? I'm not in a position to diagnose and replace all the resistors, transistors, and diodes in this amp right now (I just finished a full recap and don't have the patience at the moment).

Things change due to component drift/age, but since this bias circuit uses a fixed resistor (R5/R6) in parallel to the trimmer, the maximum resistance of the circuit is only going to be on the order of about 25Ω anyway. It would be more effective to clean/exercise/lube the trimmers as above, and experiment with something different than the 36Ω fixed resistor to alter the bias circuit overall range. A 50Ω or 68Ω resistor might allow the top end of the adjustment to get high enough to set bias correctly. Of course, you can always get to the zero ohm point by adjustment of the trimmer, so the limiter is the parallel resistance.


The left channel can be brought down to 33ish mv and the right channel 22ish mv. What long term effects would happen with the bias this high? Does it make sense to "match" the channels up to 33mv or just keep them set to minimum?

Running the bias unnecessarily high only heats the output devices more than intended, and doesn't improve crossover distortion or response.
 
Thank you for the detailed reply!

I'm taking all readings with no load hooked to the speakers, volume down, AUX mode, and all settings off. My multimeter is a Rat shack Cat 2200075 digital (was like $40-$60 new, so low to midrange but not the junkiest junk ever made), using snap in gator clips (with protective rubber sheathing) to grab the pins.

Things change due to component drift/age, but since this bias circuit uses a fixed resistor (R5/R6) in parallel to the trimmer, the maximum resistance of the circuit is only going to be on the order of about 25Ω anyway. It would be more effective to clean/exercise/lube the trimmers as above, and experiment with something different than the 36Ω fixed resistor to alter the bias circuit overall range. A 50Ω or 68Ω resistor might allow the top end of the adjustment to get high enough to set bias correctly. Of course, you can always get to the zero ohm point by adjustment of the trimmer, so the limiter is the parallel resistance.

This is a good point. It was staring me right in the face in the schematics. I'll have to dig around and find my box-o-resistors. The pots have been deoxited and cleaned, and turning them does seem to change the bias reading linearly, so maybe it doesn't make sense to replace them after all. I also noticed that I made a mistake in saying that those pots are 500 ohm anyway -- I misread the spec, they're 100 ohm. The VU Meter adjustment pots are 500 ohm.

Again, if the power supplies are affecting the adjustment, this might be a result. The same could be happening if those ultra-sensitive dual FET comparators are getting flaky.

I did a little bit more experimentation with this earlier today, and was able to get these to behave with very careful manipulation. I was afraid of manipulating them too far out of spec and relying on my eyes to get them back in the right ballpark, even to loosen them, because getting this wrong by a few degrees throws the readings off by a few volts! I did clean them up and wiggle them around by about 5 degrees, though, and they behave well enough to painstakingly move. I also experimented turning one pot while measuring the other channel, and there's not really any effect across channels.

The balance of this differential amp circuit is also going to be sensitive to the input DC voltage. Measuring the 12V supply at Q3/Q5 might reveal an unstable power supply, which would make adjustment very difficult - About as difficult as walking through Cap Square without seeing a rainbow flag. If the supply is varying on startup and then settles down, the supply itself and its connections might need some attention. A little drift is somewhat acceptable, but jumping around is not.

I'll have to check this when I get back in the machine to change out the resistors, and report back. When I had it open again earlier today, I let it stay on for an hour before taking readings instead of just ten minutes. It calmed down considerably after this. It was still oscillating by 20ish mv, though. I have no idea what level of oscillation is normal, so 20mv swing may not be an issue at all.

Just for the record, after power up, it can sway by close to a volt. It seems to start low (-400mv or so), goes high over the next couple of minutes (+400ish mv) then will slowly make its way back down to the neutral point, sometimes taking a few moments to creep back up beofre continuing on the steady drop. It seems to be happy after 30 minutes.

Again, I'll measure the power supply voltage when I get in there to work on the resistors.

---

From my work this morning, I think the preamp DC offset is at a good point. Previously I'd been being ultra cautious with my readings, not unclipping and switching my multimeter to the other channel until it was fully powered off and discharged. Today I figured since we're talking a few millivolts, it was probably safe to pull this off while the amp was on so I didn't have to wait for everything to warm up over and over. I was able to get things fine tuned much easier this way, and it's a good compromise to having two meters.

The sound is noticeably cleaner than my first go around -- about as clean as it was before I started messing with the DC offset in the first place...

I should also note that I have a donor unit with a perfectly good preamp board and working power supply (the power board and controller board on this unit are badly cracked but most of the components are good.) I'd rather not have anything blow up, but if it does, I do have replacement pieces.
 
That meter should be sufficient for most work. FWIW, if you're ever shopping for a low-cost (throw-away type) meter, I've recently become involved in informal testing a variety of imported units with varying levels of features and specifications. Of all of them I tested for function, accuracy, repeatability, and linearity, for voltage resistance, current, capacitance, frequency, etcetera, one type stood out above the others.

Calibrations01.jpg



Coincidentally, after the review reports went in, almost all the importing sellers of the "top" unit pulled their offerings at eBay and returned to the listings with the same units at nearly double the price. I'm pretty confident they all compare notes and watch the market price closely. They have since started to drop again (competition) but are not yet back in the "throw-away" price range.

The 830- and 9205-based meters were the lowest of those tested. The 9508- and 409-based meters were better, with a few types scattered between. The 890 (D series and above) were among the best tested. Of those, one stood out as having the best safety features, functions, and construction. That's the one whose price doubled in a day. I'm not going to do any endorsement or review here, but when these were well under $20 they were a steal (by comparison), but even at that you're STILL probably better off with an older/used Fluke which may be slightly out of cal:

s-l1600.jpg
 
Coincidentally, after the review reports went in, almost all the importing sellers of the "top" unit pulled their offerings at eBay and returned to the listings with the same units at nearly double the price. I'm pretty confident they all compare notes and watch the market price closely. They have since started to drop again (competition) but are not yet back in the "throw-away" price range.

Ha! Be careful with that level of power. I have the older rat-shack meter because it was the best I could afford at the time, and I needed it same day for a project (I miss that aspect of Radio Shack, being able to pick up components on the way home.) It may be time to consider biting the bullet on a higher end meter.

----

Just a quick update -- I replaced the 36 ohm resistors paired with the bias trimming pots with 75 ohm resistors I had lying around the house. It had the effect of trimming down a couple mv, but still not enough to get both channels in spec. That led be back to the trimming pots themselves.

I desoldered one and tested it. It was s-h-o-t. Completely and utterly gone out of spec. Grabbed the other one, and same thing.

As a temporary measure I grabbed the pots off of my donor unit and tested them. They had also drifted but not nearly as badly. I was at least able to get the bias level trimmed down to 18mv when paired with the 75 ohm resistors, if only just barely. I'll be putting in an order with mouser to get some shiny new ones delivered over the week. Since they're so darned cheap, I'll probably grab some at a few different resistances so I can experiment with what works best.

I can't really hear much difference with the bias set correctly (maybe just a bit cleaner), but it's running much, much cooler now. After an hour of power on, the outer case is a bit warm to the touch, where previously it would've been... not hot exactly, but quite a bit warmer.

In all the excitement, I forgot to check that to see if that 12V was stable, but I'll be back in there once I get my new trimmers so I'll check it then.
 
Quick update:

I have the KA-801 open right now. First thing I did was check the voltages on Q3/Q5. They're steady, but... far, far off spec. They're reading 24V instead of 12V. I verified this on right channel equivalent Q6, and got the same thing.

I actually wound up checking every line on Q5 and Q6. All the 11.7/12.1V lines are about double. The 52/53V lines are slightly high at around 60V. I double checked the multimeter and it's spot on.

Maybe I'm making a rookie mistake here? It seems like if things were actually that far off, the amp wouldn't be running smoothly, but everything sounds good. It also seems like if it were a bad resistor or something, it would only be affecting one channel or the other in a dual power block amp like this. FWIW The DC offset on the controller board is also bang on for both channels.

For the time being, I'm going to assume I somehow did the readings wrong and move on to the wonked out bias pot switches. I bought 100 ohm and 200 ohm replacements. I'm going to start with the 100 ohm, since that matches spec. The bias drifted since I put in those other old pots as stop gap last week (color me surprised) though at least it went towards too low instead of too high. (Probably should've just been patient and waited.)
 
Another quick update:

I've installed the new trimmers for the bias and preamp DC offset. The bias are standard 1 turn jobs, while for the DC preamp I sprung for the fancy 4 turn variety for better fine tuning.

The bias trimmers made a huge difference. I'm now able to get the bias spot on without any fiddling.

The preamp DC offset trimmers are now easier to adjust, but the pinout voltage measurements are still a bit difficult to deal with. It's still taking a considerable amount of time for these to stabilize.

A quick note on the installation of the DC preamp trimmers: the original versions had an interesting two sided construction, which allowed you to trim them through the back of the circuit board. This is the way that the spec sheet says to go about trimming them, since trimming them from the front would require a huge amount of dissassembly. I couldn't find any two sided trimmers like this online, particularly of the four turn variety, so opted instead to install compact switches on the back of the circuit board. I took care to reverse the ohm readings on the pinouts when adjusting them for installation, and I haven't noticed any problems with this installation method. They fit and work just fine.

Tomorrow, I'll be fine tuning the preamp DC offset (I don't have an hour to wait for things to warm up tonight) and hopefully putting the thing back together.
 
It will be interesting to see you results with the new pots/resistors after it has heated up and stabilized. Don't be surprised if it requires more than once adjustment after a 30 minute wait period. Temperatures will always vary, but the goal is to get it to settle after a thorough heating and adjust it as close as possible. The next time you shut it down for full cooling and heat it up again, the values may be different, but should still be within a tolerable ± percentage. That's when you are done, or you could be adjusting forever.
 
I've fine tuned everything as far as I'm willing to go. The bias and DC offset are well within tolerance. The preamp DC offset proved difficult to adjust even with the 4 turn trimmers, but after several rounds of adjustment and stress testing (ie - playing loud music with the bass turned up, loudness setting on and listening for clipping at high volumes) I'm satisfied with how everything turned out. Thanks again for talking me through some of the stickier moments.

----

Here, I'm just going to pay things forward with some information on the restore itself:

To those people thinking about working on a KA-801: the restore, including replacing the capacitors, trimmers, and other dodgy components (I'll get to that in a bit) has been transformative. It didn't sound bad before the work, but the improvements were immediately noticeable even when just pumping Pandora. There's more bass, the bass that's there is cleaner, the sound itself is more detailed, and things stay crisp at higher volumes than they did. The amplifier itself runs far cooler to the touch thanks to the new bias trimmers.

The caps are bog standard electrolytics for the most part, and slightly more expensive "audiophile grade" electrolytics in the signal path. I spent about $40 on all parts that went into this amplifier, excluding soldering / electrical equipment. I also had the extreme luxury of a donor unit, which I did use, but the parts I pulled from it were all still readily available on the market for nominal prices. The whole process took about three on-and-off weekends of troubeshooting and figuring out what actually needed to be done (initially this was just a recap and tuning trimmers, but that process uncovered other issues.)

Some of the challenges I faced:

- The four large capacitors, 10000 uf, 63V, are not available in the form factor used inside this amplifier. The only reasonably priced caps of this value that I could find were the snap-in variety. They had to be modified a bit (read: I bent the pins) to fit in the board, but I've tested them and they're quite stable where they are. If I had to do it over again, I'd probably connect them with wire leads and some sort of base to give more room to maneuver, or maybe even modify the board to accept screw in type caps.

- The donor board, which I used to take down the capacitor values, had a different revision main power board than the unit I was repairing. A handful of capacitors had to be reordered, because of this value difference.

- Another handful of capacitors had to be reordered because they were non-polarized, and I accidentally ordered the polarized versions. Be careful of this. If you don't know, polarized caps have a stripe indicating the negative terminal whereas non-polarized caps do not (I did, but apparently wasn't paying enough attention when taking down the values.)

- This amp had a peculiar issue where it would keep playing music for several seconds after being shut off, even if unplugged! Spooooky! Obviously this meant that those giant caps were being allowed to drain into the signal path, which could be bad for... everything. In my case, this problem turned out to be a failed diode, D7 on the main board, which allowed excessive current into the protection circuit HA12002, which fried out that circuit. I replaced both D7 and HA12002, and the issue went away. That circuit is luckily still available for about $5 on ebay (though I pulled mine off of the donor unit.)

- Of course, the whole point of this thread -- the trimmers for bias adjustment needed to be replaced because they were far out of spec. The VU meter trimmers, of the same build, are probably also bad but are less important. The DC offset trimmers were replaced for easier adjustment because the old ones were excessively difficult to trim. The new ones are just ordinarily difficult to adjust.

- Finally, for the preamp DC offset trimmers, I couldn't find any that matched the 2 sided design of the originals. This meant that, for practical purposes, they had to be mounted on the back side of the circuit board. Luckily trimmer pinouts are symmetrical and this can be done, so long as you remember to reverse the values from the trimmers you're replacing as the starting values. These were also tested to be quite sturdy.

As a last note, this amp is exceedingly easy to work on. I didn't have to remove any of the circuit boards to do any of this work, though in some cases this required very, very careful maneuvering around wiring. The worst I had to do was desolder some grounding resistors on the controller board and preamp board, and unhook some ribbon cables from the main board. If you're just working with the trimmers, the front and real panels are literally on rails, designed to slide out and fold down for easy access to the circuit boards. This is about as easy as it gets.

----

Thanks again!
 
Last edited:
It's been a few days, and the thought of that high voltage reading on Q5/Q6 keeps creeping back into my mind like a bad itch. I'm wondering what could cause an effective doubling.

I'm firmly in the "hobbyist" level, so it's possible that this isn't really an issue or that I made a mistake. I read DC voltage across the emitter pin on the transistor, and grounded to the chassis.

Does anyone know if this was wrong? The amp never went out of protection (likely because if the voltage that I was sending to ground.) Do I have to take the reading out of protection and ground to something else?
 
I have made until now 3 of the KA 801. The Bias Poti is normaly defekt, i change it every time. The Poti for the offset are very difficult to handle,- but it shoud work. The Service Manual is very bad, the written voltages are mostly not right, some Lead and Lag Caps are not noticed in the manual.
The third differntial stage is not realy working as it should. For that reason i change a lot of Resistors and lead and lag caps. Kenwood KA-801 schematic detail both power amps voltage amplification st....png

some of the Faults are corrected in this Schematic

Richard
 
I have made until now 3 of the KA 801. The Bias Poti is normaly defekt, i change it every time. The Poti for the offset are very difficult to handle,- but it shoud work. The Service Manual is very bad, the written voltages are mostly not right, some Lead and Lag Caps are not noticed in the manual.
The third differntial stage is not realy working as it should. For that reason i change a lot of Resistors and lead and lag caps. View attachment 1465909

some of the Faults are corrected in this Schematic

Richard

This is delayed, but thank you so much for this reply! Those values seem to be quite a bit closer to what I was seeing, and are a relief. I would think if double voltage were actually running through these transistors, I would've noticed audio issues :D

I plan on replacing most resistors in this amp with metal foils -- eventually. I've got a few other projects to cycle through first. The capacitor and bias/preamp offset trimmer swaps yielded big improvements, so I'm content for the moment.
 
This is now almost a year old, but I want to revisit it.

Over the past year, I've found myself disliking the sound of my KA-801. It seemed to be duller than I'd remembered, and I fatigued of it pretty quickly. This eventually got me doing more research on amplifiers, which eventually led me to this article: https://www.tnt-audio.com/clinica/bias_e.html

In a nutshell, the article is speaking in favor of oversetting the bias on amps made in the late 70s and early 80s. It purports that most intamps from this period in time were very conservative with their specified bias voltages, and that there were sonic benefits to oversetting that voltage (tuning the transistors in a type A/B amplifier to behave more like a type A amplifier). He suggested that with experimentation, some amps should be set to over 2x the voltage listed in the service manual!!!

I'm generally skeptical of "magic sauce" claims like this, but given that I was a bit jaded with the way the KA-801 sounded and that I had to pull it out for other service anyway, I decided to humor the article and mess with the bias voltage. You know, for fun.

Well. Hoo-lee-cow.

Oversetting this voltage made a huge difference. The sound coming out of this hunk of aluminum went from being relatively lifeless to crisply detailed with very accurate highs and bass. I can listen to it all day without being fatigued, and WANT to listen to it all day too. The midrange is still somewhat laid back, but not nearly as much as it was before. It even has some level of stereo imaging now, whereas before it was very limited.

It just sings. I'm so happy with it right now.

Keep in mind that this is in tandem with all of the other service that was performed; new caps, bias trimmers, properly set DC offset, etc. Also keep in mind that there's risk to doing things like this; I'll have to be very careful to make sure that the bias doesn't drift any higher than it already is, for instance.

In my case, I was able to get each channel up to 36mv before the amp started getting too warm for my liking. That seems to be the limit, and I'd go lower if ventilation were anything other than ideal.

But yeah, I've come full circle. I spent allot of time and effort getting that bias down to manufacturer's spec, only to find that it sounds much better overdriven. Go figure.
 
I'd like to hear if someone else also did that? i'd like to try this maybe as, i find my own 801 a bit flat to my tastes too.
 
I'd like to hear if someone else also did that? i'd like to try this maybe as, i find my own 801 a bit flat to my tastes too.

I agree with your cautious approach here. I'm not an engineer and am just trying things on my own equipment, and posting anecdotal evidence. I don't know why exactly this makes the amp sound better or the extent of the danger in doing this.

Now that I've proved that the amp does indeed sound better with the bias set high (further experimentation put the "sweet spot" where I couldn't detect any more improvement at around 30mv) I want to understand it better.

I get the general principles of an AB integrated amp, and that the bias current is essentially setting how much "overlap" there is between the pushing and pulling transistor (https://www.electronics-lab.com/article/class-ab-amplifiers/), and that this extra overlap is likely what's making the quieter passages sound better and thus improving the detail of the sound... but what I don't understand is why adding a tiny bit of extra waveform overlap between the PNP and NPN transistors would make the amp run so much hotter and pose risk to surrounding components. This is probably just ignorance of the basic operation of an AB amplifier on my part. I also don't get why the factory recommended amount of overlap would offer such an inferior sound; does this point to a problem somewhere else in the amp that can be corrected, or maybe I just got unlucky and have some badly matched transistors? Do transistors become more poorly matched as they age, and this oversetting the bias is just a mask on the problem?

Obviously, I wouldn't recommend other people do this until they do more homework. I took a risk with my own equipment to prove a hypothesis. I've proven to myself that the amp does sound better with the bias set high, but not why, and not that it's the best solution.
 
Fyi, I've started a thread in the main SS group on the subject of why boosting the bias in the KA-801 helped the sound so much, and whether or not it's safe for long term operation:

https://audiokarma.org/forums/index...ing-the-bias-made-my-amp-sound-better.945045/

The general consensus so far seems to be that it's probably safe since the heatsinks aren't getting hot, but not to boost it any more than it absolutely needs to be for optimum sound quality, and certainly no more than 40mv.

Again, this is free internet advice. Take it with a grain of salt.

That said, I'm probably going to keep mine set in the low 30s. There seem to be diminishing returns in the higher 30s, and this gives me more headroom in case values drift over time.

---

Now... if you were to do this, which I'm not recommending unless you a) have enough experience with electronics to not kill yourself on the voltages inside the KA-801 and b) are willing to take risks with your own equipment that I claim no responsibility for, you would need:

1) The service manual.
2) A good, accurate voltmeter, preferably with some kind of insulated clips on the ends of the terminals to latch onto the test points.
3) Allot of patience.

The service manual is free for download on hifiengine. It has a page dedicated to the bias and DC offsets, and will tell you exactly what test points to use and pots to trim.

https://www.hifiengine.com/manual_library/kenwood/ka-801.shtml

To overset the bias, you would do exactly as they say (using points 18 and 20 for readings on the left channel or 27 and 29 for the right and trimming vr1 for the left channel or vr2 for the right, all on the main power board X07-1640-10) except you'd ignore their recommended voltage and trim to your desired higher voltage. If I were making recommendations, I'd say to perform this in steps of 3-4mv until you're satisfied with the sound, or your millivolt reading reaches the high 30s, then stop.

You'll want to give the amp 20-30 minutes to stabilize before the reading on the bias is accurate, so this isn't exactly a quick job. It is, thankfully, pretty easy. Everything you need is right under the main cover and very readily accessible.

And, if someone else were to try this, I'd be curious to know if they had similar results.
 
Last edited:
The most accurate method is to apply an external signal, connect a dummy load, and monitor the output with an oscilloscope. Varying the input frequency from the lowest limit, usually around 20Hz, to the upper limit, usually around 15-20KHz, and adjusting bias to eliminate any zero-crossing lag or distortion throughout that range is sufficient. Any more than that is excessive and unnecessary.
 
Well...

Due to some infighting beyond my control, my other thread about why oversetting bias helps the KA-801 got poisoned and closed, so I'll continue updates here.

I've ordered a used scope off ebay (Hitachi V-355 35MHz Dual Channel Oscilloscope) and plan on reopening this beast one more time to get the actual optimal settings dialed in, plus maybe finding any other issues that might be happening with this amp.

This has been a learning experience for me, and hopefully I can add some valuable information on the KA-801 to the forums. There doesn't seem to be too much out there on this amp.
 
Back
Top Bottom