JTC111
Active member
I really hope he's talking about me.
- Status
I really hope he's talking about me.
mshilarious
Banned
Re: tube PSUs, these are vastly simpler circuits than the preamps, so there is less concern there in general, and also the DOA/defect rate on the mics is much lower. My concern is not necessarily that a working supply is dangerously defective, but that there are reports of DOA tube mics that are potentially not being correctly diagnosed. Given that some pres had half-attached power connectors (which can be very hazardous), I would open and examine the PSU on these mics, and test output voltages of the pins, before I started messing around with swapping tubes.
Again, these reports have been more elsewhere than here, so there's not a lot more I can guess/do about it . . .
Again, these reports have been more elsewhere than here, so there's not a lot more I can guess/do about it . . .
E
erniebb
New member
Totally agree. For me I haven't even got my stuff yet. And I'm no electronic repair guy either. I can totally understand the frustration that you've got something that doesn't work as well as you expected it (it could happen to me too). Yet I don't think it's fair to blame all this on the manufacturer - there're folks who are willing to buy up these "crappy" products made by this evil manufacturer if they are available. Thanks to TnC we can all be a part of this very interesting experience, but mind that the price built-in has a factor of "learning-how-to-be-a-vendor/distributor" to it. We are not buying from saltywater or musiciansfiend, and more I think of it, we the entire GB act as a distributor collectively. What I'm trying to say is, if what we are buying into has a very low failure rate, customer support to answer email/phone, good return policy, fast shipping...etc. - we probably will have to pay the 'retail' price for it.
ausrock
Well-known member
.
dgatwood
is out. Leave a message.
I'd probably check supply voltages, too, but chances are they'll decide to swap the stock tube out anyway even if there is something else wrong, so I'd say no harm, no foul.
Oh, and for your amusement, I decided it was time to own an oscilloscope. $1900 from Tektronix. Debated the models for a while before I picked that one. I figure it ought to last me the rest of my life.
Here's the power supply hum.
That's with the supply unloaded. I just bought four soft recovery diodes at Fry's. I'm going to swap them in. Something tells me one of the four diodes in the FWB is bad.
Last edited:
antichef
pornk rock
can't see the picture [and really want to...]
dgatwood
is out. Leave a message.
Weird. It's a bog-standard TIFF file as far as I can tell. Okay, I uploaded a JPEG.
I just replaced the four diodes for that 24V output stage with fast, soft recovery diodes (NTE 569) and the 24V rail hum is gone, and apparently the oscillation went with it. I'm not getting swooshing, humming, popping, etc. I was at first, but then I moved the mic away from the power supply.
For all I know, the remaining noise the last day or so may have been because the mic was too close to the power supply with the unit's cover off. Hard to say.
I'm not quite ready to declare the oscillation problem fixed... not until I've done the same part substitution on the power supply on another unit, but it very much looks like the addition of a couple of additional grounding wires along that long, narrow ground bus plus replaced PSU diodes may have taken care of the oscillation problem.
EDIT:
One more thing. I do NOT recommend the NTE 569 for DIY repairs. The part is a 600V part, which means the leads are a LOT bigger in diameter than the stock diodes. (The NTE datasheet says the body is 5mm long, but it is really more like 9mm long; if it were the size the datasheet claims, it would be perfect.) As a result, I had to drill out the holes in the board with a drill. I got lucky and didn't lift any traces doing this, but I wouldn't recommend it to anyone who doesn't know how to rebuild traces.
Also, the distance between the pins is longer for the 569, and as a result, they barely fit in the space even if you just swap out the four diodes for the EQ supply. For maximum benefit you should probably swap out all twelve diodes in this way, but they would almost certainly run into each other if you tried to replace all twelve with 569s. I just used these because I could buy them at Fry's and see if they helped. When you're throwing crap at a wall and seeing what sticks, local availability is key.
For DIY folks, there are other alternative parts that would probably be better choices in terms of fitting in the physical space. If this fix pans out with my other 81 this weekend, I'll get in touch with one of the electronics parts mail order chains and see if I can get them to stock something appropriate.
I just replaced the four diodes for that 24V output stage with fast, soft recovery diodes (NTE 569) and the 24V rail hum is gone, and apparently the oscillation went with it. I'm not getting swooshing, humming, popping, etc. I was at first, but then I moved the mic away from the power supply.
For all I know, the remaining noise the last day or so may have been because the mic was too close to the power supply with the unit's cover off. Hard to say.
I'm not quite ready to declare the oscillation problem fixed... not until I've done the same part substitution on the power supply on another unit, but it very much looks like the addition of a couple of additional grounding wires along that long, narrow ground bus plus replaced PSU diodes may have taken care of the oscillation problem.
EDIT:
One more thing. I do NOT recommend the NTE 569 for DIY repairs. The part is a 600V part, which means the leads are a LOT bigger in diameter than the stock diodes. (The NTE datasheet says the body is 5mm long, but it is really more like 9mm long; if it were the size the datasheet claims, it would be perfect.) As a result, I had to drill out the holes in the board with a drill. I got lucky and didn't lift any traces doing this, but I wouldn't recommend it to anyone who doesn't know how to rebuild traces.
Also, the distance between the pins is longer for the 569, and as a result, they barely fit in the space even if you just swap out the four diodes for the EQ supply. For maximum benefit you should probably swap out all twelve diodes in this way, but they would almost certainly run into each other if you tried to replace all twelve with 569s. I just used these because I could buy them at Fry's and see if they helped. When you're throwing crap at a wall and seeing what sticks, local availability is key.
For DIY folks, there are other alternative parts that would probably be better choices in terms of fitting in the physical space. If this fix pans out with my other 81 this weekend, I'll get in touch with one of the electronics parts mail order chains and see if I can get them to stock something appropriate.
Last edited:
mshilarious
Banned
I looked at a handheld unit once that was like $140, and I decided I couldn't afford it
Wavelab can record up to 384kHz, I should look at getting a 12 bit A/D that can do that rate. I really don't need to see frequencies above 40kHz anyway, for what I do . . .
dgatwood
is out. Leave a message.
I looked at a handheld unit once that was like $140, and I decided I couldn't afford it
I bought a handheld for $300 first. I got it home and found that it only went down to .5V per vertical division, so all the hum I was looking for was basically a pixel tall on the screen and was lost in the noise of the crappy circuitry anyway. So I returned it and bought the Tektronix.
The only other handheld scopes they had were either A. made by the same manufacturer as the crappy one (with better vertical resolution, but still a lousy, very limited UI/feature set) or B. one that I couldn't try out by a manufacturer I've never heard of.
So that basically left me with two options: spend $3-400 for a tube-type that weighs as much as my entire rack or spend a minimum of $1,200-ish for a Tektronix that I can move around easily. Since my electronics workbench consists of a cardboard folder in random spots on the floor of my TV room, portability is pretty important to me.
So I pretty quickly ruled out the 3000 series since they write data to floppy disks (can you even BUY floppy disks anymore!?!), which narrowed it down to two 1000 series models and three 2000 series models. One of the 2000 series models was an old version that wrote to CF cards and was half again more expensive, so I quickly ruled that one out as well. In the end, the color display on the four-channel unit won me over, both in the ability to easily distinguish between the different inputs (for the rare occasions where I need to do so) and in terms of generally being easier on the eyes. I decided against spending an extra $400 for moving up to 200 MHz bandwidth. I figure I'll never need more than 100 MHz even doing occasional video stuff.
So after much deliberation, I picked the TDS2014B. Amazingly, Fry's price was actually competitive. I think that may be the first time this has ever happened.
Anyway, it's pretty sweet. The probe cables are shielded with a ground wire hanging off near the tip (which is a push-clip type tip), which means it rejects all sorts of environmental noise that really would make it horrible for an unshielded probe. About the only concern is that I'm going to have to buy a bunch of small USB memory sticks while you can still buy them. For whatever reason, these things will only write to media that is 2GB or smaller (buggy FAT16 implementation?). Hopefully they'll fix the firmware someday---there's no good reason for any limitation when you're dealing with USB devices other than programmer sloppiness. (I say this with confidence, having looked through enough FAT filesystem code to have a pretty good idea what's involved.)
Last edited:
W
wreckd504
New member
After reading the last 5 pages - I'll take you up on that. I've got two 81's that haven't shipped from Chances yet. Check your PM's.
ausrock
Well-known member
Hey Wreckd,
No-ones saying anything yet, but there's a number of guys on our local mastering forum that have pre's on order and I bet some of them are feeling just a little edgy.
P
P. Glaser
New member
Thanks. I guess I need to open it up and take a look.
I have a 310 and a 900. The PSU's are the same chassis.
antichef
pornk rock
ouch! -- the sting of ignorance - it hurts! I know you're planning to look more into this over the weekend, but if I could hazard a question: So far, I've only been able to locate "fast, soft recovery" diodes at that 600V rating. I see "fast recovery" diodes ("fast recovery rectifier" diodes, that is) at lower ratings. How important is the softness?
dgatwood
is out. Leave a message.
I'm not certain, but I get the impression that the softness is key to avoiding the high energy spikes that aren't getting filtered out sufficiently by the caps. Newark has a wide range of parts, but they don't keep them in stock in the U.S. For example:
http://www.newark.com/vishay-general-semiconductor/ug1b/diode-fast-recovery-1a/dp/98K2125
or perhaps:
http://www.newark.com/vishay-general-semiconductor/sbyv27-200/diode-soft-recovery-2a-200v/dp/26M2922
If my testing this weekend confirms that this change is a useful and necessary fix, I'll see if I can get them to stock a few thousand units stateside.
antichef
pornk rock
OK - that makes sense - when a diode blocks current going the wrong way, it has to "recover" before normal operation resumes (major oversimplification alert goes here, I bet), and a "fast recovery" means it does so quickly and a "soft recovery" means it does so quietly (because normally, it might produce a lot of electrical noise) -- both are important here. I guess.
crazydoc
Master Baiter
I don't think switching to "soft recovery" will make any difference (though I admit to complete ignorance of soft recovery diodes - I believe they may be used in switching power supplies to reduce EMI.) If you're finding 20+ mv of ripple in the supply (if I'm reading it right), I'd think it's due to failure of the regulator, not the diodes or filter caps.
mshilarious
Banned
Especially when you consider all the audio gear in the world that uses bog-stock 1N400xs for rectifiers, and doesn't hum like bumblebees
antichef
pornk rock
Would it make sense to switch out the regulators (eg, LM7812,LM7912 -> LM317,LM337)? Big circuit change?
edit: oops - dumb question - looks like the 24v rail is already using LM317? That's from the schematic earlier - I don't have my units in front of me. The 12 volt is using the 7812/7912. I'm a be quiet now. But if replacing the regulators is a good idea, I'm all ears.
Last edited:
mshilarious
Banned
Check the voltage feeding the regulator on a stock unit--I wonder if somehow the drop is larger with the stock diodes, causing the LM317 to go out of regulation
G
gemsbok
New member
any step by step info for the transistor swap on the 81-s.
- Status