EMRFD Message Archive 851

Message Date From Subject
851 2007-06-11 18:55:27 w1kilofoxtrot Excessive current drain in Analog Signal Processor?
I'm testing the analog signal processor from section 9.10. I've
been calling the project the T2Pro, for short. I'm seeing higher than
expected current in the 5532s. I'd expect around 30mA from the 12V
supply for the five devices, resulting in a 3V drop over the 100 ohm
decoupling resistor. I'm seeing more like 75mA, so the supply voltage
to the op amps is down to 4.5V which is only just barely in operating
range.

Both channels have similar DC voltages, and I can't find any
shorts. The first two stages, mike amp and buffer, seem to be working
fine, but as I plug in more op amps to complete the filter/phase
shifter, the currents climb to sort of unmanageable.

Are there any 5532 experts out there who can suggest some other
things to check? It seems unlikely I got 5 bad ones, but they are from
the same date code.

The amps are not oscillating as far as I can tell; the scope traces
on my Tek 475 look clean.

Thanks for any tips. I can't wait to get into the upconverter...
That's the next stage to tackle!

73,
Steve
W1KF
853 2007-06-12 04:52:38 Allison Parent Re: Excessive current drain in Analog Signal Processor?
854 2007-06-12 17:56:30 w1kilofoxtrot Re: Excessive current drain in Analog Signal Processor?
>
> You either have something oscillating or a short. If you using
> sockets shorts should be checked. It's also possible that an
> output is hard against the rail.
>
> While your at it check the 100ohm resistor to be sure it really is
> 100 ohms.
>
> If it's none of that I'd suspect you have a few bad 5532s.
>
> Allison
>

Well, neither oscillating or short, actually. I still don't know
exactly what was going on, but removing and reseating the amps in the
sockets brought the current down to about 40mA. Still higher than I'd
like to see, but the board is working.

Definitely not oscillations, though I've heard the 5532s are
susceptible to that issue.

And, not an output short, at least, since the amps all still work
properly.

Maybe an open input? Not sure how that would affect Idd, but I may go
ahead and replace the sockets.

Well, on to the upconverter. I'm starting to think I'd like to
simplify the design with a low noise, low impedance op amp instead of
using the external driver. I'll probably try Rick's design first, but
that's an interesting future development.

Thanks for the thoughts!

73,
Steve
W1KF
855 2007-06-12 18:56:30 Allison Parent Re: Excessive current drain in Analog Signal Processor?
857 2007-06-13 07:11:59 w1kilofoxtrot Re: Excessive current drain in Analog Signal Processor?
>
> Lower the value of the 100 ohm resistor if the voltage drop
> concernes you. However the signal amplitudes are low and large
> output swings are note required. This is more important for a
> reciever then TX (RX needs higher dynamic range).
>
Yeah, I can lower it, but it's fine now. The Vcc will be high enough
to be within spec.

> > Definitely not oscillations, though I've heard the 5532s are
> > susceptible to that issue.
>
> Just about any opam can, 5532s are fairly tame.
>
> > And, not an output short, at least, since the amps all still work
> > properly.
>

> Those opamps are output protected against shorts. More than likely
> you had an input floating.
>
Must have been brain fog. I thought I'd read the 5532s would be
damaged by output shorts. Must have been when I was building the R2Pro.

> > Maybe an open input? Not sure how that would affect Idd, but I may go
> > ahead and replace the sockets.
>
> Over the years (35 at last count) I've been burned badly by sockets
> unless they were of the machined pin types. I rarely build using
> them now as component quality is good and rare is the need to
> replace something like an opamp.
>
> FYI the most common failure with socket is a pin that curled under
> and makes intermittent contact with the socket.
>
I'm thinking that my home etched board may have contributed to a bad
solder joint, maybe. The tin plate helps, but there's nothing like a
plated thru hole for security.

> > Well, on to the upconverter. I'm starting to think I'd like to
> > simplify the design with a low noise, low impedance op amp instead of
> > using the external driver. I'll probably try Rick's design first, but
> > that's an interesting future development.
>
> You need an opamp with a good high current output and they tend to
> be fussy. Ricks design is bulletproof, I've used that scheme as line
> drivers to 10mhz with good results every time and as audio drivers
> to fairly high power. The only thing to watch for is getting enough
> standing current but, not too much. A few mA usually does well in
> avoiding crossover distortion. The circuit in EMDRF Figure 9-83 is
> what I assume your planning. The noise figure is set very early in
> the system and most any opamp will be low noise as the gain required
> is low. The real worry in this last stage is not noise but distortion
> as you pushing a low (50ohm) impedence load and you desire a constant
> and stable driver impedence.
>
Amen to that. I think I may just have the ticket, though. National has
an ADSL driver, the LMH6672, which is very low distortion, wideband
opamp, designed to drive 50 ohm loads! I'd need to do a gain study
overall to make sure I can get the output level I need to the mixers.
I think the magic number is around -10 dBm, but I need to confirm
that. I'll want to get the current board working first anyway, in case
anyone wants to duplicate it.

>
> Allison
>
>
>
And let me repeat:
> > Thanks for the thoughts!
> >
> > 73,
> > Steve
> > W1KF
> >
>
867 2007-06-26 07:22:36 w1kilofoxtrot Follow up: Excessive current drain in Analog Signal Processor?
Well, it was stable for a while, but now it's singing like a crazed
chickadee, at 2 MHz. I've bypassed each device, to no avail, so I'm
getting a bunch of 10pF caps to add across the feedback resistors.

Yes, they -were- in the schematic, but since I had built the R2Pro ASP
and had no problems, and used virtually the same layout, I thought I
might get by. No luck. Perhaps it's to do with the date codes on the
5532s, but I'm not taking the R2 apart to see.

When it's stable and working, I'll report back.

BTW, the mic amplifier works great. I've propped the mike in front of
the speaker
868 2007-06-27 07:10:20 Allison Parent Re: Follow up: Excessive current drain in Analog Signal Processor?
869 2007-06-27 15:06:51 w1kilofoxtrot Re: Follow up: Excessive current drain in Analog Signal Processor?
--- In emrfd@yahoogroups.com, "Allis