Aston Origin crackling noise troubleshoot

[SamsaGregor]

First of all, "Leg 1 / 2 / 3" is very arbitrary. Second, in-circuit measurements of transistors are rarely conclusive.

In this case, the 726 (i'm guessing, mV?) you measured between collector and emitter (middle pin and left pin, going by this last-attached photo) would be the forward voltage of the (zener) diode in parallel with the transistor.

Re-measure those voltage drops keeping in mind they're supposed to be PNP's, and see what you get.

That being said, even if the diode drops look "sane", that doesn't say anything about how noisy the transistors may or may not be.

If you can't get your hands on a pair of 2SA1015's, 2N5401's will work just as well - just mind the pinout!

Alright, got it! I have a cheap multimeter and for some reason it didn't display any symbol in diode mode, but yes, I think it's mV. You are correct; it was tricky to find 2SA1015, so I will probably have to go for 2N5401. Since the measurement doesn't say anything about the noise, I might as well replace them, as this looks to be the main culprit, just to get it out of the way?

I have one of these with the same behavior but it's been on the back burner for a while now since I've got other mics. I guess it's time to pick up this project and we can sort it out together since we might have the same problem. I remember when I first started looking into it that this was a common complaint and I think Aston was doing replacements at some point.

Mine has some embarrassingly bad soldering and lots of flux residue so I'm going to start there:

View attachment 138357


I have a spare LSK170A, two KSA1015, and two of the one gig resistors on hand already. I don't the have the SMD tants, though.

That's a great idea! I'm thinking about ordering the transistors, the FET, and capacitors, just in case. I will probably not make much progress until then, but I will update here in the thread.

I actually emailed Aston twice asking for advice, but I didn't get any response back on either occasion. I wasn't really expecting any help, but it was worth a shot.

 

[Khron]

I might as well replace them, as this looks to be the main culprit, just to get it out of the way?

Sure, just be aware it might not be the cure. In my sE X1, it must've been some resistor - replacing capacitors / diodes / transistors did nothing.

 

[jaymz168]

I just fired up mine to test and I get an increased noise floor with the pad inserted and it sounds like it's picking up AM when the high pass is turned off. So I proceeded to clean up the board with some 99.9% IPA, DI water, and a brush. I use a Datavac ESD Duster to blow dry PCBs but I'm still going to wait a bit to power it up and see if it's behaving any better. I tried to work some Deoxit D100 into the switches and cycled them about fifty times each but I'm not really sure anything got into there. I'll probably touch up some of these joints as well before I put it back together.

Hopefully that does it but if not I'm going to go down the road of replacing components. I know when I built some Mic Parts mics the instructions were very clear not to touch the gigaohm resistors with your bare hands so hopefully just cleaning it up will help. The levels and impedances at the front end are non-trivial and super sensitive.

Before cleaning:


After cleaning:

 

[Voyager10]

Regarding the transistor measurements, I was mostly asking for the DC voltage (with respect to ground) of the transistor pins, when the mic is powered. The output transistors should have similar voltages on their equivalent connections; if they don't it might tell us something,

It might be worth getting a big (e.g. 47uF, 50-63V) capacitor and soldering it in parallel with each of the other big capacitors on the board; power it up each time, and if the crackling has gone or gets better, the cap is worth replacing. Might save some time compared to repopulating the whole board.

 

[jaymz168]

I'm just waiting on some 10uF radials to pursue this further, I only have axials on hand apparently. Cleaning it up didn't have any effect on my noise floor or the radio pickup. It did help with the bit of switch noise that I had though.

 

[ccaudle]

Cleaning it up didn't have any effect on...the radio pickup

If you are testing with the PCB out of the case I think that may not be unusual. If it is back in the case when testing several people have reported problems when the mesh protecting the capsule became loose and did not make secure electrical connection to the rest of the case.

 

[jaymz168]

If you are testing with the PCB out of the case I think that may not be unusual. If it is back in the case when testing several people have reported problems when the mesh protecting the capsule became loose and did not make secure electrical connection to the rest of the case.

I did fully reassemble it but I'll check continuity on the headbasket.

 

[SamsaGregor]

Hello, I just wanted to give a quick update. I have not disappeared or abandoned the project. I have been waiting for some components. I ordered caps, transistors, and a FET. I am crossing my fingers that they'll come pretty soon since I will be going away by the end of next week to do an internship for a couple of months. But I'll update you as soon as I know more!

 

[Somkereki]

I have another question about the transistors. The markings on them are A1015 YCG48. When I measured them, I had the positive in the middle and the negative on either 1 or 3. That would make them NPN, right? But when looking for replacement transistors, A1015 are PNP. What am I missing here?
Also, from what I understand by the voltage drop, leg 1 is the emitter and leg 3 is the collector?

It is important to clean up the flux residues! With isopropyl or ethyl alcohol. It has caused me many problems. Even in untouched, factory devices. If you need to replace a transistor, I would recommend the 2SA 970 pnp 2SC2240 npn transistors. (Low 1/f noise) It has given me the best results so far in microphone applications. Many people don't care about z 1/f noise, they say we don't hear it, because below 10 Hz it can cause significant but serious intermodulation, which can be heard!

 

[jaymz168]

Hi all, some updates from here:

• These mics apparently have a reputation for poor QC and noisy capsules from the factory. Mine is a very early one so there's a good chance that my capsule is just bad. It's difficult to get a good picture of it but there's some spots on the diaphragm and some general color variation going on. When I got this mic some friends were doing vox at home and I offered this to use. It ended up being noisier than an SM58 so I ended up just using it as a desktop gaming mic for a while. Now I'm going through some project backlog and decided to sort it out.
• The following measurements have all been done with mic powered for at least 30 mins prior to allow it to settle thermally.
• To help narrow things down I replaced the capsule with a capacitor. The capsule measures at 63.1pF @200kHz (should I measure at 1kHz instead?) so I subbed a 65.7pF polyester cap which is the closest value that I have on hand. If I understand correctly this should simply move the peak of the response a bit right?
• 
• I measured this "mic" with REW through a MOTU M4 with phantom applied and 0dB gain. The results are the same with all combos of pad and hpf. I've built some mics kits before but I'm not familiar enough with circuit level condenser stuff to confidently interpret this but my gut says this looks fine. However the averaging is hiding the intermittent noises that are the real problem.
• The thing that you can't see/hear in the graphs is that the noise changes. If I listen to the signal from the mic with the dummy "capsule" in place I can hear a couple noises that are variable and are the same as the noise with the capsule in place. For the most part it sounds like that graph: it sounds kind of like being on a passenger jet with that low mid peak. But there's a broadband noise that comes and goes. It's sounds like it's tilted towards the low frequencies so maybe it's flicker noise? Maybe something heating, resistance changes, and current changes? There's also a popping noise and then and the popping nature implies shot noise to me but I'd love to hear other suggestions. Maybe something charging and discharging. Here's an short clip recorded with 60dB gain that has a bit of the variable noise :
• My setup is far from anechoic in either the acoustic or RF domains but I can say that doing stuff on my phone while I hold it near the mic produces no change so I do think the noises are internal, not external. Same with doing stuff on the laptop and plugged vs unplugged: it doesn't affect the noise at all.
• I already have a Quantasylum QA403 which can give me absolute measurements in dBV, etc. unlike the MOTU but it has 50R inputs and no phantom power so I've gone ahead and ordered a QA472. I also emailed Aston about a replacement capsule and got an email back from Music Group so I guess the Borg has swallowed them up too They pointed me towards Audio Lab of Georgia for parts but I think I'm the first person to call them about one of these becuase they said no one has ever asked for a replacement capsule before. I'm waiting to hear back about pricing and availability
• Given these factors I may try to slap a third-party capsule in and mod the board values to match it. The QA472 should be super helpful for this.

 

[BIASKING]

I've found that cleaning all the excess flux from the PCB can reduce spurious noise.
The pics show a lot of flux on the board.
A swab and some anhydrous alcohol works pretty well, and easily available.

 

[Vladimir_Lesnyak]

Я сейчас попробовал ваши предложения для контактов XLR. При подключении только контактов 1 и 3 я получаю очень много шума. Шум похож на шум, который обычно издает микрофон (неприятный треск). Я не слышу свой голос через микрофон.
Если подключить только контакты 1 и 2, то, говоря относительно, я получаю очень мало или вообще никакого шума, и это не то же самое. Я также могу слышать свой голос через микрофон.

Я также проверил транзисторы мультиметром в диодном режиме. В схеме с отключенным микрофоном; надеюсь, это правильно. Однако мне понадобится помощь в чтении значений.
Транзистор 1: Ноги 1 и 2 считывают 767. Ноги 2 и 3 считывают 762.
Транзистор 2: Ноги 1 и 2 считывают 770. Ноги 2 и 3 считывают 762.

Следующее, что нужно сделать, — это проверить трюк, применив тепло, как предложил Сомкереки.

​

Hi. I think the 47 Ohm resistor or capacitor in the 47uF+20-47 Ohm series circuit that connects the 3rd passive XLR contact to the common negative of the circuit is faulty. If it is a Shoeps circuit, then check the resistor going from XLR3 to the output transistor emitter.

 

[Vladimir_Lesnyak]

The value of the mentioned resistor may be different, for example, about 20 Ohm. If an accidental short circuit occurs between XLR3 and XLR1, then the smd resistor often burns out from the discharge of a large-capacity electrolytic capacitor charged to the Hi voltage 30-40V.

 

[jaymz168]

I've traced my microphone PCB and uploaded a schematic for the "Issue 2" pre-MT version of this microphone to the Technical Documents forum. It's mostly complete, some small cap values are missing because I can't get a good read on them in circuit. Schematic is here : https://groupdiy.com/threads/aston-microphones.89398/

I've ordered a replacement capsule from the repair shop that MT pointed me to, Audio Lab of GA. I think it's going to be a long while before I get it since they didn't have any in stock and even though I placed this order two weeks ago their site still shows as zero in stock, zero on order, and zero due from China.... Their site says they place one big order each month so we'll see what happens. I don't think the capsule is entirely problem for me because the low freq noise is present with a polystyrene capacitor in place instead but it doesn't look great to begin with.

I've gotta say, tracking down noise on this is a pain lol. It has to be assembled or the environmental noise swamps the self noise which makes this all take much much longer. The soldering on this PCB generally does not look great so I may just redo about all of it.

@Vladimir_Lesnyak Thanks for the pointer, unfortunately I don't have many SMD on hand. I think it must be time to buy a book of them. The resistors in question are indeed 20R. Thin film should theoretically be quieter and I can get closer tolerances for hopefully better CMRR, do you think 1/8W thin film would be ok with the expected inrush current?

EDIT: I've updated the schematic, some of the pad section was incorrect. I still don't know the value of the caps in the CLC filter as they all measure as open on my Fluke 117 and Peak LCR meter. I tried removing one to test and of course I squeezed the tweezers too hard and it went flying into oblivion.... I have spare 10n and 100n 0805s here anyway.

 

[Khron]

do you think 1/8W thin film would be ok with the expected inrush current?

The "inrush current" will already be limited by the pair of 6.8k resistors the phantom power gets fed through, within the preamp, so power rating is of no real consequence.

 

[jaymz168]

The "inrush current" will already be limited by the pair of 6.8k resistors the phantom power gets fed through, within the preamp, so power rating is of no real consequence.

That's what I suspected, thanks.

 

[Voyager10]

It looks like the pad and HPF switches on the schematic should be swapped (i.e. SW1 should be labelled Pad, and SW2 labelled HPF).

One very simple test would be soldering a wire across the C13 terminals, so the FET stage is powered down. If the crackle disappears it's something to the left of
SW2, if it stays the same it's SW2 or something to the right of it.

 

[jaymz168]

It looks like the pad and HPF switches on the schematic should be swapped (i.e. SW1 should be labelled Pad, and SW2 labelled HPF).

One very simple test would be soldering a wire across the C13 terminals, so the FET stage is powered down. If the crackle disappears it's something to the left of
SW2, if it stays the same it's SW2 or something to the right of it.

Man, I've gotten so mixed up flipping this tiny PCB back and forth trying to follow the traces and vias. Good catch, thank you!

I'm going to try your suggestion today and see if that leads me anywhere.