• 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

Kegger/blueglow kt88 project

Chassis question: For those of you who have used polished aluminum top plate for chassis, what if anything do you use to keep the aluminum from showing corrosion? Apparently, using 2024 plate from a scrap metal yard may not have been the best idea. After a little research it seems that 2024 is far more corrosion prone than whatever allow is normally used form chassis construction. It is very strong and stiff but really corrodes easily. I made a nice-looking oiled walnut frame, but the aluminum mounting plate doesn't look good. I suppose it doesn't really matter since is will live in my wood shop and no one but me will notice it.:(
 
powder coat. Sorry couldn't help it as it doesn't really answer your question I think but it's my favorite solution. If you want to leave it bare, a coat of Boeshield is made specifically to keep corrosion from forming on metals. I use it on my table saw top and nothing showing after about 2 years. It leaves a very thin waxy coating. Can't really feel it. Or you could use some auto paste wax.

aerosol version:
https://www.amazon.com/Boeshield-Ae...hvlocphy=&hvtargid=pla-4583932713646657&psc=1

pump version:
https://www.amazon.com/Boeshield-RF...hvlocphy=&hvtargid=pla-4584070153202127&psc=1

John
 
I thought about powder coat and even looked up a local service. However, it was over a holiday and they were closed. I really wanted to start wiring, so I cleaned it with solvent and used rattle-can lacquer. I've used that on wood a lot, but I've found the hard way its not good for aluminum. It was humid when I sprayed. It looked good on a test piece and the real chassis when I first sprayed. Since I had it turned upside down for wiring I didn't notice that "blushing" had occurred and I had developed white spots in several places. I am now in "disaster recovery" mode!

I use bowling alley wax on my table saw and band saw table and it works very well. Should have just waxed the aluminum and stopped.
 
I use self etching primer from the local hardware store on bare aluminum before painting with your typical rattle can lacquer. But I live in a dry climate, typically 10% to 20% humidity.
 
kward said:
I use self etching primer from the local hardware store on bare aluminum before painting with your typical rattle can lacquer. But I live in a dry climate, typically 10% to 20% humidity.
Click to expand...
It was probable 90% humidity here in Southeast Virginia when I sprayed. I have a dehumidifier in my shop and can usually keep it down to 50%. I did not realize that humidity when spaying was a problem. I typically use oil finished on my wood projects and had no experience with lacquer on metal. Now I do!
 
Clipping question:
With approx. 1.3vrms input on the driver ECC85 driver grid, I get about 10.5watts rms into 8 ohms on the speaker terminals just before clipping. The clipping is always on the bottom half of the sine wave first. Is there any reason to try adjusting bias so that the clipping symmetrical? By that, I mean should it start at the same point on both upper and lower half of sine wave.
 
JohnB1950 said:
Clipping question:
With approx. 1.3vrms input on the driver ECC85 driver grid, I get about 10.5watts rms into 8 ohms on the speaker terminals just before clipping. The clipping is always on the bottom half of the sine wave first. Is there any reason to try adjusting bias so that the clipping symmetrical? By that, I mean should it start at the same point on both upper and lower half of sine wave.
Click to expand...

It should, but almost never is perfect. If the other side start clipping within a reasonable output, I wouldn't bother.
 
And even if you were to tune it so it clips more or less symmetrically, the increase in power output would likely be negligible relative to what your ear can perceive as an increase in loudness.

That assumes that it’s already pretty well tuned and that the other side is not far from clipping at the point the side you noticed starts to clip. So if you’re getting ~10 watts out as is, you are already quite close to theoretical max power output anyway.

On the other hand, if you want to fiddle for the sake of experimentation, perhaps just to understand how the amp behaves, then by all means tune it up to your liking.
 
Thanks again for the input. That is what I suspected. For me, it is a learning thing. I've been reading all I can about biasing, loadlines and such and am finally just starting to grasp. I have been playing around with a couple of online load line calculators just to see if I can duplicate the operating conditions for the ECC85 driver in the Blueglow/Kegger schematic.
 
If you're getting a solid 10 watts out of the amp I think you're there in terms of tuning/power. Mine shows the same asymmetrical clipping as you describe but it's not severe. The other end of the sine wave starts to clip right after the lower end.

When i first looked at it, mine was clipping way harder on one side. It turned out not to be an output tube issue at all. Going back through the circuit I came to realize that the driver tube was not properly biased. It was running too cold and the lower half of the waveform was running into cut off. After fixing this the output clipping became much closer to symmetrical. I don't know if you will ever get to true symmetrical clipping in a SE amp. I see the same issue when working on SE guitar amps. One side always clips before the other.
 
All single ended amps have asymmetrical clipping. As the grid voltage becomes large and negative we eventually reach a point where the anode current goes to zero which is the cut-off. The only way to alter this behavior is to chose a different operating point either by adjusting the bias or the load. However, my assumption is that this may then increase harmonic distortion since the operating point is usually chosen to minimize non-linearities in gain when the tube is operated relatively far away from cut-off and close to the operating point.
 
I am calling it finished for now until I hook up my horn speakers and listen for a while. Sounds great with my cheap inefficient bookshelf speakers. I tweaked the ECC85 cathode resistors a little and ended up with 150ohms. That gave me the highest output before clipping by a very little. That gave me 6.1ma cathode current and 2.5v bias on the driver tubes. Plate dissipation on the KT88s is just under 35watts which from what I read is about max. This was after dropping the input from 121vac to 117vac and that gave me 448vdc on KT88 plate and 245vdc on the ECC85 plates. For the cathode bypass cap I used the one Mark suggested in his BOM which was a Nichicon KZ Muse. Would a higher quality film cap help any for sound quality in that position? My input power circuit with relays is programmed to keep from cycling power no sooner than 2 minutes after turning off and bypassed the thermistor after 30 seconds so that it is already cooled down and ready for next power-up. This was probably overkill but fun to do. I just noticed I need to reverse the chock bell cover bolts to the heads are forward-facing.

IMG_1294D.JPG
 
Congratulations on finishing the build! I used the same bypass caps you did and I'm pretty sure a high quality film cap will not make much, if any, difference in that position. If it was a coupling cap you would definitely want film.

I am almost finished testing the Aikido phono preamp I've been building to go with this power amp. It's been a great learning experience.
 
I'm sharing a pair of speakers between this amp and my AVR. I'm changing to a relay for the speaker switching that will have this amp as the normally closed connection but just as a backup, incase the relay doesn't make a solid connection, should I put a resistor across the speaker terminals? What ohm?
 
Some time ago I looked at the designs for switching between two (or more) guitar amplifiers and one set of speakers which is basically the same scenario as you are dealing with. You may find more information by looking at guitar head switchers. If I remember correctly the way that many of them work is to switch the amp between the speakers and a dummy load in which case you need a 8 or 16 ohm power resistor of the appropriate wattage. That doesn't go across the speaker terminals but is switched into the circuit when the speakers are connected to different amp. In guitar amp rigs it is assumed that all the amps are on all the time so each amp must see a load all the time (either the speakers or the dummy load). Obviously, if you turn both amps off before doing the switching this issue doesn't arise unless you forget to do so. Many of the commercial products incorporate a design in which the dummy load is switched between the speakers and amp just before the speakers are disconnected from the amp in order to make sure that the amp always sees a load rather than a momentary open circuit as the relay opens or closes. This may be overkill for your application.
 
After running my amp for three hours continuous, the Edcor XPRW-033 PT is sitting at 130 degrees Fahrenheit. Is this what everyone is getting with this transformer? I potted it and used anti-vibration and have absolutely no hum or noise from the PT. No hum at all electronically either. After listening the rest of today, I plan on lugging it downstairs and out to the shop where it will live. This sucker weighs in at 42 pounds!
 
I generally think that if I can keep my hand on it continually it's well within/under limits. Some PTs run HOT!
 
Don't get too wrapped up around the voltages. A few volts difference in line voltage at the house can cause that. Heck, I get different line voltage readings in my house from minute to minute depending on the load (dishwasher, dryer, HVAC, etc.) in play. You are fine here.

JohnB1950 said:
I can't seem to get a grasp on Mark's published voltages. If my ohm's law is correct, the circuit from the B+ to the ECC85 plate goes through the 4.7k and the 39k resistor in series. They should have the same current. Mark shows 460v B+ and 442v after the 4.7k resistor and 229v after the 39k resistor. That would be a current of (460-442)/4700=3.83ma through the 4.7k resistor. The current through the 39k resistor should be (442-229)/39000=5.46ma. They should be the same current unless the 200k feedback resistor is doing something I don't understand. In any event, I plan on tweaking the 39k plate resistor a little to try to get the cathode current in the ECC85 down from my 6.5ma to Mark's 5.5ma and see what the plate voltage is at that condition. I am probably over-thinking this whole thing. The readings I are will within specs for the ECC85 but not for some of the other tubes that can be substituted. If I start doing any tube rolling may worry more about the voltages.
Click to expand...
 
You can chase that but if you are getting 10.5w, no need. If you were clipping at 4W, then yes. Nice looking build BTW.
JohnB1950 said:
Clipping question:
With approx. 1.3vrms input on the driver ECC85 driver grid, I get about 10.5watts rms into 8 ohms on the speaker terminals just before clipping. The clipping is always on the bottom half of the sine wave first. Is there any reason to try adjusting bias so that the clipping symmetrical? By that, I mean should it start at the same point on both upper and lower half of sine wave.
Click to expand...
 
You must log in or register to reply here.
Back
Top Bottom