[BUILD] NU FEDERAL AM-864 PCB OFFICIAL BUILD & HELP THREAD

[bernbrue]

Which output transformers do you use? How did you wire it? The 12ax7 could be bad as well. Can you measure the voltages around the output tube once again? Has your input transformer got a  centertap? Check for correct wiring.  Compare with the data sheet of your transformers.
Bernd

 

[fiat_cc]

I used the Edcor  XSM 10K/600 output transformers as recommended on the first page of the thread.  Input I used  Hammond 140QEX 1:1 600ohm transformers.  The Hammond has both sides centre tapped, so I connected it on the output side of the input transformers, and just blanked it off on the input side, as there is nowhere to connect it to the board.
Output transformer I connected the centre tap on the primary side (TO-OUT-TX on the board), and left it disconnected on the  secondary (FROM-OUT-TX) connection.

The bit that is annoying me, is whatever the issue is, it's the same issue on both channels, because they both have the same hum. To me that kind of rules out a bad tube.

 

[bernbrue]

The circuit itself is absolutely flawless. The pcb has been tested as well. So there must be a very simple, obvious mistake. Please double check the wiring of the XLRs and verify that hum is not introduced to the unit from an external source.
Bernd

 

[fiat_cc]

I'm not suggesting at all there is an issue with your product.  I'm just frustrated, as I feel as though I did a good job of building this, and I cant figure out where the hum is coming from. I'm not a tube guru, but I've built my share of tube circuits successfully, and feel as though I have a sound understanding of layout issues etc.  I'm always learning of course, but I thought this one would fire right up with no issue. 

I will partially disassemble and try a few things in terms of layout etc, and triple check my wiring.
Regards,
Reuben

 

[scott2000]

Maybe a leaky cap?

 

[fiat_cc]

Maybe a leaky cap?

Not impossible, but given that it's a 2 channel unit, and both channels exhibit virtually identical noise issues I'm leaning towards something that is shared between both boards.  I have some sheet aluminium at home, and will try building a shield of sorts around the PT.  I've just been dealing with a guitar amp build that has all sorts of noise issues due to the OT placement too.  I'm feeling it's either a layout thing, or possibly something like meters.  I may have made a wiring mistake on both boards, but I was careful with this build, and it seems unlikely.

I've put it on the shelf to concentrate on another project for a few days, and will come back to it with fresh eyes soon.

 

[shot]

Hi!

I'm in the process of finding what's wrong with my one channel build.
I'll explain in the next post what is wrong and ask for assistance if I don't find the error in the meanwhile.

But I think I found an error on the pcb silkscreen!
It's in the section of output pad.
Labels on R23 and R19 are reversed.

I've drawn myself a diagram  of actual situation on the left and schematic on the right.
If you look at top left on my diagram you'll notice that R19 is connected to output pot CCW (that's supposed to be R23 on schematic) and R23 is connected to output transformer (that one is supposed to be R19 on schematic).

If you stuff the board following silkscreen labels you end up with wrong values. It works of course, but it's wrong!

I attached my hand drawn diagram (sorry for the lack of drawing skills  ???)



Luka

 

[zayance]

Yes you're correct shot, silk error there.
I surely have crossed the names while cleaning the board's silk...

Sorry for the inconvenience, will add a note for this error.

T.

 

[shot]

Unfortunately, I'm struggling to find free time to work on this compressor more.
I don't have a clue what's happening with compression. I don't hear it!
Meter is showing that it's compressing, attack and release slightly change the way needle on the meter is moving, but when pushed I only notice more distortion. I don't hear the compression. Os should I say, I think I hear some compression but it may even be placebo when simultaneously watching the meter bounce.
Input potentiometer is definitely not behaving as it should since whatever I do it keeps the same level. I went to check if it's connected the wrong way but it seems fine. I'll double check. Maybe that's the root of all my problems.
I'll keep you posted as I go along when I get some time to work on it.
Btw, I have 10k:10k as input transformer and at the moment I'm using 10k:150R as output. I know that this output transformer may not be appropriate but I don't think it's got anything to do with my problems at this stage.



Luka

 

[bernbrue]

Hi Luca,
You might check the video I made from my unit and compare the compression you clearly can hear there with yours. Attack and release changes are subtle with this compressor.  I mentioned this before. Your input pot should cut down the input signal to zero when turned fully counterclockwise, whereas the output pot is just a 10dB attenuator.  Try to get the input pot wiring right, Im shure this is causing your problems.
Heads up!
Bernd

 

[fiat_cc]

Well, I rebuilt the power supply end of things and made a shield for the transformer out of light gauge galvanised steel, and the problem persists.
I’m going to strip the chassis, re examine boards and solder joints, rewire the whole front panel, and try to track down the issue.
I’m wondering about ground issues. There are two points on the board, one labelled ground, and another labelled PE with a ground symbol. I have both these points on each board running back to a single shared earth. Am I inadvertently creating a ground loop by doing that?

Cheers,
Reuben

 

[bernbrue]

No, the grounding seems correct. Indeed there are two points on each pcb, GND and PE, which need to be wired to a common chassis ground nearby the IEC inlet (verify that there is a solid connection to the base plate of the enclosure). This chassis ground needs to be connected to the IECs  main ground.
XLR input and output pin 1 need to be connected properly as well. Check attached picture ones again.
Cheers
Bernd

 

[fiat_cc]

I’m currently away on work again, but I’ve measured to the ground pin on the mains plug, from each ground on the pcb and am well under 1 ohm everywhere (closer to zero ohms).
One thing I did do was populate the 9v psu section on both boards, but I don’t imagine that would creat hum.
I’ve ordered new 7 pin tubes, in case there’s issues with  the old NOS tubes I originally got.

I plan to strip the whole chassis (besides the power supply which I just rebuilt and shielded), and see what I can find. It’s gotta be something causing it, so I just need to find what it is.

 

[fiat_cc]

I may have found my issue, in the heater filament wiring, so I wanted to ask a question (seeing as I now have my unit totally stripped down).  In every tube amp I've ever built or serviced running twin triodes like 12AX7's on 6V heater voltage, the filaments are always wired where pins 4 and 5 are tied together, so pin 4/5 get one side of the heater wiring, and pin 9 gets the other.  When I assembled the board, I couldn't get continuity from the terminal block to pin 9, so I hard wired that connection.  Upon deeper examination, I can see that you have pin 4 +, pin 5-, and pin 9 apparently has no connection that I can discern, so am I right in thinking you are running the heaters in series, rather than parallel?  If so, doesn't that make the heaters very under powered at only 3.15V per filament, or am I missing something in the way that you've put this together?  In any case, the way I had wired it is not in keeping with the way you have designed the board, and therefore is quite likely the cause of my hum, but I would like to understand what we have going on here with filament wiring before I reassemble everything and find out that wasn't my issue (pretty sure it is though).
Regards,
Reuben

 

[bernbrue]

The heater supply is  6,3 VAC with 100 Ohm resistor to ground. The wiring is in parallel. Have a look at the picture of the heater wiring in the first post and the picture attached.
Bernd

 

[fiat_cc]

Hi Bernd.
The attached schematic shows heaters in series.

Each tube might be in parallel with each other tube, but a 12ax7 has two triodes, each with a 6.3v heater. If you run them in series (I.e. pin 4 and 5, with pin 9 disconnected) it requires a 12v supply (hence the 12 in the tube designation), if you connect pin 4 and 5 together, and connect one side of the heater supply to pins 4/5, and the other to pin 9, the 6v heaters are in parallel and run off a 6v supply.
I’m sure that the way you have the board made is running the 12ax7 filaments at 3.15v each.

 

[fiat_cc]

In fact, just doing some reading and found an article about running 12a-7 series tubes on starved heaters. Apparently a 12ax7 will still conduct at 3.15v heater voltage but only produce about half the gain making it similar to the output of a 12au7 running at full heater power. I am guessing this is why you couldn’t get satisfactory results out of the 12au7 that you originally designed around as being a close replacement for the original Federals 6SN7GT.

 

[bernbrue]

I was talking about all tubes being wired in parallel. Indeed the 12AX7 is wired in series and this could have been the reason why the 12AU7 didn’t produce enough gain. Nevertheless to solve your hum problem you should do the heater wiring as shown in the first thread.
Bernd

 

[fiat_cc]

I’m going to go ahead and jumper pins 4 and 5, and hard wire my filament wiring to pins 9 and 4/5 and try a 12au7 in there. Obviously your design works, but is running the tube a long way from the tube manufacturers operating parameter specifications (tube manufacturer datasheets say not to go lower than about 6v).
I’ll let you know results.

 

[fiat_cc]

I’m yet to run audio through it, but voltages are better. Cathode voltage on the 12au7’s is around 5.3 volts, which is much closer to the bias suggested in the tube manufacturers  specs sheet. Spec sheets suggests grids at -8v for a plate voltage of 250v and grids at 0v for plates at 100v, so +5.3v cathodes (grids at -5.3v) for plates at almost exactly 200v seems to be well within an acceptable range.
All other voltages measure exactly as per the original measurements for this unit, so I feel this is a step in the right direction.
Now I just need to make sure that the hum has gone, and it passes audio as expected.