[lordearl]

Quote:

I use my MiniMAX with a pair of uFonken speakers featuring Fostex FF85WK drivers, which are 86 dB efficiency. This works will for me in a near-field setup (sitting at my computer). I had asked TomB about this before I tried it, and he advised me to watch that it didn't get too hot. I have not had any heat problems though. I use the high-gain tubes anyways because my main headphones are Sennheiser 580's.

 
Thanks GSPENCE - great news.
Did you change the output capacitors in order to lower the bass frequency response?

 

[zwack]

I got it working now 

. Thanks again. DB and tubes are biased and it's now running at 27V. The only thing that doesn't work is the relay circuit. I did a quick search and came across another thread started by someone who had the same problem.
 
I'll quote some of your (tomb) posts from that thread.
 

Give us a voltage measurement across DM2 - the 1N4148 diode. It should be close to 12V. If it is, then yes - the relay is bad. I kind of doubt that, though.

By the way - I forgot that we couldn't see your resistors because of the way you turned them. Any way you can confirm the value of RM1 and RM2? 

I measured the voltage across DM2. It measures 3.3V instead of 12V. Something is obviously wrong here, but I can figure out what..
 
 

1. Measure the voltage at the point indicated on RM1 relative to GND:

This will confirm that you have full voltage from the power supply going to the relay circuit.

It measures 27V, so full voltage going to the circuit.
 
 

2. Next, measure the voltage across two points indicated on DM1 as shown:

This will confirm that the Zener diode is the proper part and is in the circuit to drop 12VDC.

The voltage across DM1 is 11.5V. The Zener must be doing it's job.
 
 

3. Next, measure the voltage at either of the points indicated on DM2/RM2 relative to GND:

This will confirm that at least 12VDC is getting to the relay.

My DDM measures 10.9V. Should be enough to switch the relay, but for some reason the relay doesn't switch.
 
Here is a image of my relay delay circuit (click to enlarge):

 
As you can see I used the BD139 for QM1. I installed it per the directions on the Minimax relay-delay page. QM2 is not revealing itself on the photograph, but I can confirm it's a MPSA14. 
 
The next photograph shows the values of the resistors and one can even read the text on the Zener (thanks to my camera and .. not to forget the operator 

). All values are as per the BOM. 

 
I'm certainly not an engineer, but my best quesses are that the 1N4148 is working like it's supposed to be or... the relay might be faulty.
 
Edit: forgot to mention that I waited for over a minute to start measuring.

 

[tomb]

You didn't show the bottom of the PCB.  It might be that you have some cold solder joints on the relay.  Can you try re-flowing the solder on the relay pins?  The first two next to DM2 might be the key (they're the coil connection).  Be careful of the heat, though, it may be just as likely that you could've burned out the relay coil in soldering it.  I think it may be the former rather than the latter, though.
 
If you are measuring close to 12V across the leads of DM2 (you said 10.9V), then there's no reason the relay should not actuate.  That voltage will vary from very low to 12V, as the capacitors charge.  As long as you wait more than ~45 seconds, you should be measuring the steady-state voltage.  If the coil at the two pads next to DM2 are registering close to 12V and the solder joints are good, then the relay must be bad.
 
According to the data sheet, the relay coil should actuate at about 8.4V maximum.  That means that it may actuate at an even lower voltage.  So, if you get 10.9V and you do not hear a "click" at about 45 seconds (cold start) or 30 seconds (hot start), then either the solder joints are bad or the relay is.
 
P.S. Once you hear a click, that may not be all there is to it.  It could be the Left and Right signal contacts are not connecting into the circuit on the PCB.  One thing at a time, though ...

 

[zwack]

I just did a test with another 12V relay I have lying around. I put it on a breadboard, put a diode over the coil and connected the negative side to ground from the Minimax and the positive side to the 12V side of DM2 (Where the text "DM2" is printed onto the PCB). The relay actuated just fine. Than I connected the negative side to the other side of DM2 (just like the relay on the board) while keeping the positive side connected to the same point. Now the relay refused to actuate. It looks like the voltage difference is too small between the positive and negative sides of the relay's coils. 
 
I can more or less confirm this when I measure the voltage over DM2 directly from power up. It starts to rise slow to 3.1V. Than it stops rising and stays there. 
The voltage between the negative side of the coil and ground is 7V! This explains why the relay won't actuate. Now to find the source of this problem.
 
ps. I have not checked for bad solder joints yet, but I do not think they are the source of the problem. To be sure I'll re-solder all components from the relay-delay circuit.

 

[zwack]

PROBLEM SOLVED! 
 
After some time studying the relay-delay circuit, the only thing that I could come up with was that QM2 was not functioning like supposed to be. It was not going full open or something like that. That would also explain why the voltage difference between the negative side of the coil and ground was 7V. QM2 did not short it to ground. The MPSA14 (QM2) is now replaced by another one and the relay is working fine now. 
 
I always get a bit anxious when things don't work, especially when I don't fully understand what's going within the circuits. Still leaning and determined to get there
 
Now to test the amp with some music. I will report on how it goes and eventually post some pics of the finished amp.

 

[tomb]

Quote:

PROBLEM SOLVED! 
 
After some time studying the relay-delay circuit, the only thing that I could come up with was that QM2 was not functioning like supposed to be. It was not going full open or something like that. That would also explain why the voltage difference between the negative side of the coil and ground was 7V. QM2 did not short it to ground. The MPSA14 (QM2) is now replaced by another one and the relay is working fine now. 
 
I always get a bit anxious when things don't work, especially when I don't fully understand what's going within the circuits. Still leaning and determined to get there
 
Now to test the amp with some music. I will report on how it goes and eventually post some pics of the finished amp.

Way to go!  Glad to see you got it working!

 

[zwack]

And there is music! I temporary hooked it up to my computer using some old RCA connectors, just for testing. Sound is a bit thin, but that could well be coming from the computer's soundcard. When I finished the build I will be using my left-over Rega DAC in combination with my CD-pro transport. I have a good feeling that the Rega-CD-pro combination will sound better.
 
I still have a question about the tubes:
 

1. I'm currently using 12FM6 tubes. At the time I ordered them I was not aware of the tube matching service at Beezar. The ones I got are from Raytheon & Tung-Sol and rated 90 and 85  Whatever that stands for. On both boxes "RCA Side HS" written, assuming that they where produced in the same factory. Are the tubes close enough matched?

 

[tomb]

Quote:

And there is music! I temporary hooked it up to my computer using some old RCA connectors, just for testing. Sound is a bit thin, but that could well be coming from the computer's soundcard. When I finished the build I will be using my left-over Rega DAC in combination with my CD-pro transport. I have a good feeling that the Rega-CD-pro combination will sound better.
 
I still have a question about the tubes:
 

1. I'm currently using 12FM6 tubes. At the time I ordered them I was not aware of the tube matching service at Beezar. The ones I got are from Raytheon & Tung-Sol and rated 90 and 85  Whatever that stands for. On both boxes "RCA Side HS" written, assuming that they where produced in the same factory. Are the tubes close enough matched?

Probably.  "Millett" tubes were only made by four manufacturers: GE, RCA, Tung Sol, and Sylvania.  So, no matter what the original tube box says, or the painted-on tube label, it's one of those manufacturers that made it.
 
As for the notations on the top of the Beezar tube boxes, I've always intended to document this thoroughly, but it's one of those things that you never quite get around to doing:
 
Top Left - tube box or tube label mfr
Top Right - tube tester reading
Middle - tube designation
Bottom Left - "real" mfr
Bottom Middle - getter type
Bottom Right - plate type (graphical representation)
 
"RCA Side HS" - (I'm assuming this was on the bottom) means RCA was the mfr and the getters are Horse-Shoe type mounted on the side of the plates.
 
Unfortunately, we need to state that you didn't order a kit.  Tubes are completely matched according to test, mfr, and construction with tubes that come with a kit.  I try to do my best if someone orders tubes separately and they don't pay for matching.  In the case of 12FM6's, those tubes are very scarce.  Beezar may represent the largest stock in the world right now, but that's still only about a hundred.  In your case, the painted-on labels for the tubes have little to do with the actual mfr.  Not seeing/remembering the boxes specifically at this point, it sounds like I gave you two RCA-manufactured tubes with identical construction that are 90 and 85 tested output.  That's within 10%, so they should be pretty good.
 
Not to be outrageously self-serving, tubes are not like opamps.  Every one may sound different.  You are encouraged to buy as many as you like (not just from me) so that you can experience the differences.  You never know when you might find that golden tube that does everything just as you want it to.

 

[zwack]

Thanks a lot for the explanation. 
 
After I became aware of the tube matching service I ordered two matched 12FK6 tubes along with some other stuff I still needed. These tubes are both rated at 75 and do come from the same manufacturer. I will try these after some time to get used to the 12FM6 tubes. This will make my ears more receptive for the differences between the different types of tubes. If it shows that I prefer the 12FM6 tubes over 12FK6's than I will surely order some more from you and make sure that I'll go for the matching service!
 
I'd like to share that I'm very satisfied with the way you handle your orders and the support you give. 

 

[tomb]

Quote:

Thanks a lot for the explanation. 
 
After I became aware of the tube matching service I ordered two matched 12FK6 tubes along with some other stuff I still needed. These tubes are both rated at 75 and do come from the same manufacturer. I will try these after some time to get used to the 12FM6 tubes. This will make my ears more receptive for the differences between the different types of tubes. If it shows that I prefer the 12FM6 tubes over 12FK6's than I will surely order some more from you and make sure that I'll go for the matching service!
 
I'd like to share that I'm very satisfied with the way you handle your orders and the support you give. 

Thank you for the very nice comments!

 

[zwack]

Currently listening the the third album using the MiniMAX. I'm very pleased with it, but compared to the left channel the right channel is just a tiny bit louder. I'm using the 12FM6 tubes with are not matched (left: 85, right:90). Can the small volume offset be the result of the unmatched tubes?

Ps. All transistors used are bought matched from Beezar.

 

[Beefy]

Could be the pot as well, could be wax in your ears.
 
Run a 60Hz sine wave through your source/amp. Measure the AC voltage at the input, after the volume pot (try a few different volume levels), and at the outputs. Most multimeters will pick up 60Hz AC just fine, and this will tell you where any large channel imbalance might be.

 

[tomb]

Or, you could just swap the tubes from one channel to the other.

 

[zwack]

Quote:

Could be the pot as well, could be wax in your ears.
 
Run a 60Hz sine wave through your source/amp. Measure the AC voltage at the input, after the volume pot (try a few different volume levels), and at the outputs. Most multimeters will pick up 60Hz AC just fine, and this will tell you where any large channel imbalance might be.

Definitely no wax in my ears. Reversing my headphone also reversed the louder channel 

. I ran a 60hz sine wave through the amp and this revealed that the AC measured (at the output) at the right channel was indeed higher. Measuring at the input before and after the pot revealed that the pot's channels are perfect in sync. Thanks for the advice. 
Quote:

Or, you could just swap the tubes from one channel to the other.

After the measurements I swapped the tubes and measured again. Now the difference was a lot smaller. Than I realized that I had to adjust the bias again. After adjusting the difference was still a lot smaller that with the tubes reversed. When listening the difference is now hardly noticeably.  
 
Is it possible that the channel imbalance is caused by the diamond buffer stage, which is somewhat corrected by the unmatched tubes (or emphasized when reversed)??? Or maybe I'm just talking nonsense??  Is there a way to check my theory?

 

[tomb]

Quote:

Definitely no wax in my ears. Reversing my headphone also reversed the louder channel 

. I ran a 60hz sine wave through the amp and this revealed that the AC measured (at the output) at the right channel was indeed higher. Measuring at the input before and after the pot revealed that the pot's channels are perfect in sync. Thanks for the advice. 
After the measurements I swapped the tubes and measured again. Now the difference was a lot smaller. Than I realized that I had to adjust the bias again. After adjusting the difference was still a lot smaller that with the tubes reversed. When listening the difference is now hardly noticeably.  
 
Is it possible that the channel imbalance is caused by the diamond buffer stage, which is somewhat corrected by the unmatched tubes (or emphasized when reversed)??? Or maybe I'm just talking nonsense??  Is there a way to check my theory?

No - tubes are imperfect devices.  People try to treat them like opamps, but it's sort of the difference between a horse and mule* (tubes).  You can actually compensate for differing tube outputs somewhat by adjusting the bias by ear.  Get both tubes as close as you can in measured bias, then pick one and turn the trimmer screw with your headphones on until you you're satisfied the sound is centered.
 
Trouble with that is that the inherently stronger tube is still going to respond more strongly to music peaks.  You can't really hope for much more than a 1-2 dB difference.  They're always going to be slightly off, but you can get them close enough that you won't notice with the proper tubes.  Keep in mind that some Millett tubes will continue to drop in bias for 1-2 hours.  Plus, it's always possible that you haven't run these enough.  Bias may change radically with NOS tubes used for the first time in decades - they get little bits of gas in them from sitting so long and it takes some running in to burn out those bits.
 
I think you've ordered some more, so perhaps out of a greater selection you'll get something closer to what you want.

 
 
* Just a caveat to that - mules are often stronger and better than horses in many things, it's just that they don't automatically pop out of the chutes ready to run.  You have to adjust with them a bit, learn their traits, etc.  Once you do, you may find that they do many things better.  Well, anyway ... it's an analogy, nothing more.