Built the Pearl, having problems. Please help

[Saurav]

Hi,

I finished the Pearl last night. It sounds great, but there are some bugs I still need to work out, and I'm in way over my depth here. I'd really appreciate it if someone could help me out. I'll put down the facts that seem relavant to me, and if I leave out something important, please let me know.

* I built the Pearl almost exactly to instructions. I braided the umbilical from the power supply to the main chassis (mostly for mechanical strength), in case that's an issue. The umbilical has 5 wires, the 5th connects chassis ground of the two enclosures. PS and signal outputs join at the chassis ground of the main enclosure, then go back to the PS enclosure through this 5th wire (I plan to experiment with this to see what gives the least hum).

* My linestage is a transformer-based passive. The input impedance at DC measures around 60 ohms.

* DC voltages check out, mostly. All the numbers marked on the PCB are within 5% except for the 11.7V, which is 10.7V for me.

* Wiring dress - power supply wires inside the main enclosure are twisted and zip-tied close to the chassis. Signal wires are pretty much direct from solder pad to RCA.

* My Pearl is picking up a radio station. This, obviously, is RFI. I've never had RFI problems with a phono stage before.

* Along with the radio station, there's a squealing or howling sound. This changes in frequency by itself sometimes. It changes drastically if I touch or move either of the input wires (RCA to PCB).

* I don't have a 'scope, but I think I can hear some distortion. Bass transients at high volume definitely distort.

* It's built in a basic Hammond aluminum box, the top cover is currently off, and the box has holes in the corners (i.e. isn't fully enclosed). Placing the cover on the box didn't make much of a difference to the problem.

* If I turn the volume down and touch the input RCA on one channel, there's a 'pop', and the squealing stops. Any further movement of the input cable doesn't bring the squealing back, and the obvious distortion on bass attacks also disappears.

* On the other channel, the RCA connector pops every time I touch it. The squealing goes away on this channel too, but the noise floor is higher (this channel has an LED on it, so I'll investigate the noise floor later).

* Everything runs more or less fine until I disconnect the Pearl, solder something in and plug it in again, then the squealing is back until I touch the RCA and make it pop.

Those are all the observations I can think of. Questions:

* Is all of this RFI? Should I just use shielded input cables, and that'll solve all my problems? Or is something oscillating, which would then probably be a much bigger problem, I guess.

* There's an output coupling cap, so the near-short at DC that my linestage presents shouldn't be a problem, right? (With a previous phono stage my woofers started pumping the moment I'd select the phono input on the linestage). I tried a 200 ohm and a 5 Kohm resistor in series with the output, but that didn't seem to make much difference, maybe the squealing went down a little bit.

* Can a bad solder joint cause RFI pickup?

* I can't think of much else. I guess I could take it in to work and borrow a 'scope from one of the hardware labs, but I really wouldn't know what to look for.

Any help would be most appreciated. This is the first component I've tried to build that didn't come with parts and connect-the-dots instructions, and it's a little frustrating to be thrown into the deep end like this

Thanks a lot,
Saurav

 

[GRollins]

Cure the phono stage?
No problem.
First, you slaughter a goat under a new moon. Spread the entrails...
Oh.
You meant something that involved a soldering iron.
I'm not sure but what you don't have more than one problem. Try any or all of the following:
--Reflow all your solder joints. Yes, a bad solder joint can give you RFI. Pay particular attention to the solder joints on the RCA jacks and where the signal meets the board. Just because the solder joint looks okay, doesn't mean it made a good connection. I've got a buddy who loves tinkering with things, but can't solder for diddly-squat. It scares me to look at some of his solder joints. Any time we go into one of his pieces of equipment, he always begins with,"Now, Grey, don't fuss..." After which, I fuss.
--Check the shielding on your cables. If your shielding has gone kaput, you're going to get all kinds of noise. I blew out some irreplaceable tweeters one time when the ground lifted on my interconnects. Not a good feeling. Be careful.
--The DC resistance of the transformer isn't so much the question. What we need to know is the AC impedance. If it's 60 ohms, I'm assuming there's a fair amount of wire in there, so the impedance is probably fairly high.
--Braided wires can give you horrendous amounts of capacitance and inductance (hence, setting you up for evil things). Since the wire is coming from the power supply, we'll assume that it's carrying DC when there's no audio signal...still, you might try unravelling the braid, just for the sake of argument. At least it's a cheap option.
--Play with the dress of your internal wiring. Sometimes wires like being near the chassis. Sometimes not. You did ground the chassis box, didn't you?
--Grounds. Ground the turntable to the phono stage. Then not. Ground the phono to the line. Then not. Try every possible permutation of ground/not ground. Do not neglect grounding one or more of the above to a known-to-be-good earth ground. Do not assume that the ground lug on an AC outlet is an actual ground. Check it.
--Pour yourself a libation, sit back and look the situation over as calmly as possible (this is after you get past the tearing-your-hair-out stage) and rethink everything. Remember to forgive yourself ahead of time, because you're going to hate yourself after you find the one, silly, idiotic thing that you forgot to check earlier.

Grey

 

[Saurav]

Thanks, that makes me feel better

Reflow all your solder joints. Yes, a bad solder joint can give you RFI.

I already did the input and output RCAs, since one of the wires caused a scratchy sound when moved. I'll check the board too. While on that topic - this is my first time with a "plated-through" (I believe that is the term) PCB, and on some joints the solder flowed through to the screen-printed side of the board. Is that normal? Are those my good joints or my bad joints?

Just because the solder joint looks okay, doesn't mean it made a good connection.

Learnt that with the last preamp I'll build. I didn't know bad solder joints could cause RFI issues, I'm more used to pops and crackle and scratchiness.

I've got a buddy who loves tinkering with things, but can't solder for diddly-squat.

Sounds like me

Check the shielding on your cables.

There isn't any. At least, not upstream from the phono stage, nor inside the phono stage (TT is a Rega P3 with the stock RB300 cables). I'll see if touching/moving the phono cable has the same effect. I didn't think about that since I've had that TT for over a year now, and I haven't faced anything like this. I'll check the solder connections on those RCAs too.

If your shielding has gone kaput, you're going to get all kinds of noise.

I get much more noise when I leave the input RCAs open than when the phono cable is attached, so I'm guessing that my phono cable is fine. I'll double-check it anyway.

The DC resistance of the transformer isn't so much the question. What we need to know is the AC impedance.

I thought it 'reflected' the input impedance of the amps, multiplied by the square of the turns ratio (in step-down mode)? I'm not sure what the primary inductance of that unit is, in case that's the figure you're talking about. My CD player and tuner drive it fine, and my previous phono stage drove it fine too. The $25 Radio Shack "Little Rat" oscillated.

Braided wires can give you horrendous amounts of capacitance and inductance (hence, setting you up for evil things).

Can give, or do give? That would be a bad idea for interconects then, right? I was thinking of trying out something along those lines.

Since the wire is coming from the power supply, we'll assume that it's carrying DC when there's no audio signal...still, you might try unravelling the braid, just for the sake of argument. At least it's a cheap option.

-Play with the dress of your internal wiring. Sometimes wires like being near the chassis. Sometimes not. You did ground the chassis box, didn't you?

There's a terminal strip screwed on to the chassis box, and a wire from the middle (grounded) post goes to my DIN connector, and thw wire in the DIN connector connects to the chassis of the PS box and the green wire from the power cord. Does that sound right? It means my main chassis isn't grounded if the DIN is disconnected, but I didn't see a way around that.

Ground the turntable to the phono stage. Then not.

An RB300 is grounded to one of the RCAs, so I can't experiment with this.

Thanks a lot for all the help, this really means a lot to me (corny as that sounds).

Saurav

 

[Saurav]

Follow-up:

Well, re-flowing the solder definitely helped. I also re-routed some of the grounds. Now the squealing and radio voices seem to have disappeared completely (though there seemed to be a little present when I first fired it up), and the ground loop hum has become worse. That's much easier to debug though, and I think I have some idea of what to do there.

Thanks once again, and I may come back with more questions if something else seems to be wrong

Saurav

 

[Saurav]

Well, people still seem to be reading this, so I'll post another update. I think most of it was just plain old RFI pickup on the input cable. I rigged up a makeshift "shield" - wrapped copper tape from the hardware store's lawn/garden section (meant for keep snails and slugs out of a garden) wrapped around PVC tubing from the plumbing section, and that took care of all of the squealing and radio voices. I've ordered "real" shielded cable. Grounding changes have cut the hum down too, but that needs more experimentation, there's room for improvement there.

 

[GRollins]

Grounding the turntable via the RCA jacks works well in many cases, but it doesn't give you the option of <i>not</i> grounding the turntable. I have seen cases where the turntable motor was throwing an EMF field (not the case here) and it was worth running a separate ground from the turntable just to drain off the electrical motor noise--it was contaminating the ground in the preamp.
Sounds like you're making progress, though. Keep at it.

Grey