Ultrasonic cleaner...

[Joe Gwinn]

On Thu, 30 Jul 2020 17:17:53 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:

On 7/30/2020 4:53 PM, Joe Gwinn wrote:
On Wed, 29 Jul 2020 17:30:51 -0700, Don Y
blockedofcourse@foo.invalid> wrote:

I was hoping to avoid probing the transducers directly. I figured I
could isolate the HV and just verify that the *drive* to it was
working properly. I.e., assume transformer and chokes are intact
so if driven correctly but still no apparent output power, the
transducers must be toast.

I don\'t think that not probing the transduced is a successful
approach. This is a power circuit - not too fussy. Just cob togther
a voltage divider.

I figure if the primary is being chopped, a lack of output power is
either:
- bad xformer
- bad chokes (all 3!)
- bad xducers (all 3)
In either case, I won\'t dick with it any further. (these sorts of things
turn up fairly regularly so its easier just to wait for another to come along)

The frequency will be 30 KHz to 50 KHz, most likely. Won\'t be MHz at
that power level, I\'d guess.

What is the make and model of your unit, anyway?

https://www.ebay.com/itm/VWR-Model-150D-Aquasonic-Ultrasonic-Bath-/143178745430

Apparently it has a different \"part number\" than \"model number\"
so chasing it down on the VWR web site was tedious.

WVR no longer offers this brand and model, it seems. But they ought
to have repair information. I\'d call them up on the phone and ask.
Also, see if they will tell you who made the unit for VWR.

There was no AC power transformer, so an isolation transformer is
essential for working on such a unit. Then, you can ground whatever
part of the circuit is convenient.

There are at least two xformers (IIRC) in the unit. But, one is undoubtedly
for the controls (MCU/display) and the other likely for the transducer step-up.
ISTM that if the entire unit was behind a transformer, that would likely
be pretty (physically) large -- i.e., I should remember having seen
something like that.

Exactly.

Displays blink, buttons seem to be recognized... just no \"bzzzzz\". So, I
assume the controls are working (with possible exception of the interface
to the HV) and the problem lies in the HV section.

I think I have a 500VA isolation transformer lying around from the days
of designing KWHr meters. That\'s probably about the rated size of this load
(4A @ 120VAC).

I have such a transformer. Tripp-Lite? One must open it up and
disconnect the green ground (and isolate the spade terminal with
plastic tubing).

\"Triad-Utrad\". And, apparently only 250VA.

Mine is a Tripp-Lite unit, also 250 watts. I paid $100 in 2006 for a
new unit at the local electronics parts store.

They also make a 500-watt unit. I bet the same outfit makes these for
both brands - only the label varies.

The trick will be accessing the electronics without having to do
a major teardown (the insides look like they are pieced together
from lots of \"modules\" instead of designed as an integrated whole)

But many more points can be accessed, if only you knew what all those
wires were.

Yes, but it\'s a mess in there. Like it was pieced together instead of
being designed as an integrated unit. Like it was designed by someone
in *Detroit* (motor city)!

\"Control processor\" is mounted behind the indicators -- indicators mounted
on the solder side of the (thru hole) board. \"Power\" board mounted on
the base -- with a case-mounted heatsink supporting the switching Q\'s
jutting out over much of the electronics. I.e., if I want to probe stuff,
I will have to remove the boards while staying close to the xducers due to
lead lengths as well as the Qs on the heat sink.

And, of course, ensuring that there\'s water in the tub in case the
thing actually tries to start!

My plan is to trace out the HV section and see how it is gated on/off.
Then, verify the presence of that gate when commanded by the controls.
Then, go looking at the chopper to see if there\'s any signs of life, there.

You can also cob together a optically isolated HV probe with a
resistor string driving the LED (with anti-parallel diode protecting
against reverse voltage) in an optocoupler (with at least 2500 V
withstand), and measure the photocurent from the coupler photo diode
or transistor. This will allow you to see if the voltage is remotely
plausible.

Joe Gwinn

 

[Don Y]

On 7/31/2020 10:58 AM, Joe Gwinn wrote:

The frequency will be 30 KHz to 50 KHz, most likely. Won\'t be MHz at
that power level, I\'d guess.

What is the make and model of your unit, anyway?

https://www.ebay.com/itm/VWR-Model-150D-Aquasonic-Ultrasonic-Bath-/143178745430

Apparently it has a different \"part number\" than \"model number\"
so chasing it down on the VWR web site was tedious.

WVR no longer offers this brand and model, it seems. But they ought
to have repair information. I\'d call them up on the phone and ask.
Also, see if they will tell you who made the unit for VWR.

I found this:
<https://www.nist.gov/system/files/documents/ncnr/UltrasonicCleaner_VWR.pdf>
but it only deals with usage. And, the details of the control panel
(e.g., visual depictions) differ. Perhaps an earlier/later version?

There\'s also a discrepancy in \"with drain\" as mine clearly has an integral
drain while the manual claims the 5.7L model does not. <shrug>

[What the hell is \"degassing\"? Operating the transducers at a lower
power level??]

I can try calling them -- but that might take the \"fun\" out of the
exercise -- turning it more into a \"job\". Likely next week as this
week is effectively over. (But, part of the appeal of rescuing things
are the puzzles that they present and the challenges of solving them)

I think I have a 500VA isolation transformer lying around from the days
of designing KWHr meters. That\'s probably about the rated size of this load
(4A @ 120VAC).

I have such a transformer. Tripp-Lite? One must open it up and
disconnect the green ground (and isolate the spade terminal with
plastic tubing).

\"Triad-Utrad\". And, apparently only 250VA.

Mine is a Tripp-Lite unit, also 250 watts. I paid $100 in 2006 for a
new unit at the local electronics parts store.

They also make a 500-watt unit. I bet the same outfit makes these for
both brands - only the label varies.

I am more and more convinced that I had a 500 when I did the KWHr meters.
I\'m going to assume -- if I had one -- that it is probably packed away with
the prototypes. And, that this smaller one is a more recent acquisition
(\"rescue\").

Regardless, they are likely identical internally.

(This one has a single receptacle on a pigtail; IIRC, the \"other\" one
had the receptacle built into the enclosure/case.)

The trick will be accessing the electronics without having to do
a major teardown (the insides look like they are pieced together
from lots of \"modules\" instead of designed as an integrated whole)

But many more points can be accessed, if only you knew what all those
wires were.

Yes, but it\'s a mess in there. Like it was pieced together instead of
being designed as an integrated unit. Like it was designed by someone
in *Detroit* (motor city)!

\"Control processor\" is mounted behind the indicators -- indicators mounted
on the solder side of the (thru hole) board. \"Power\" board mounted on
the base -- with a case-mounted heatsink supporting the switching Q\'s
jutting out over much of the electronics. I.e., if I want to probe stuff,
I will have to remove the boards while staying close to the xducers due to
lead lengths as well as the Qs on the heat sink.

And, of course, ensuring that there\'s water in the tub in case the
thing actually tries to start!

My plan is to trace out the HV section and see how it is gated on/off.
Then, verify the presence of that gate when commanded by the controls.
Then, go looking at the chopper to see if there\'s any signs of life, there.

You can also cob together a optically isolated HV probe with a
resistor string driving the LED (with anti-parallel diode protecting
against reverse voltage) in an optocoupler (with at least 2500 V
withstand), and measure the photocurent from the coupler photo diode
or transistor. This will allow you to see if the voltage is remotely
plausible.

I pulled the unit out of the ToDo pile last night for a closer look.
It\'s going to be tedious to service, given that it won\'t want to
be run dry. The \"buttons\" are part of a membrane keypad which is
adhered (obviously) to the exterior of the enclosure. This puts
limits on how far the \"control board\" can travel wrt to it. I.e.,
I likely won\'t be able to access the control board with the transducers
under risk of being powered (e.g., if, by chance, there\'s a flaky
connection that my probing repairs!). Maybe I can hack together
a stand to lift the assembly up while maintaining the \"holds water\"
orientation so I can reach in from below... (?)

Or, put the unit into a bucket, though upside down, open the drain
(to let the trapped air vent) and then let the water level in the
bucket SURROUNDING the unit rise to just \"above\" the xducers (which
means the bottom of the \"inside tub surface\" contacts the water)??

But, regardless, I can remove the boards and carefully trace what\'s
going where to get a feel for how the circuit is intended to operate.
Lots of \"loose (though connected) wire\" inside but not very effective as
service loops.

Then, decide if I can hack together a placement that allows me to power
the unit with the boards still \"out\". Else, reinstall them and work from
there.

[Will anything likely go poof if I just disconnect the wires to
the xducers? Or, do I need something to load down the HV outputs?]

I\'m presently out of bench space as I have several \"live\" prototypes
staged there that are all internetworked. Plus, all of their field
connections. So, I\'d like them close enough together that I can probe
any of them -- possibly triggering from one of the others -- without
having to spread out into separate rooms. I have an upcoming dog-n-pony
and I\'d like to get as much functionality running, as possible.

This means finding some space that isn\'t *intended* for \"my\" use and
(temporarily!) commandeering it. The living room floor is a favorite
of mine but \"she\" tends to get annoyed if the \"mess\" lingers for
more than a few hours! (and Heaven Forbid I burn the carpet with
a careless soldering iron placement or solder splash!!)

The dining room table is equally problematic! (do we *really* need
to SIT, there, to eat??)

Kitchen counters? Ha! Don\'t even THINK about it!!

Sheesh! What *silly* priorities...

[Jeff L\'s \"shop\" sounds pretty good, now!]

Eventually (i.e., just before the demo), all of this gets stuffed *into*
ceilings and walls so it\'s out of sight. But, debugging places greater
accessibility on the criteria!

[Maybe I can stake out some space in the garage and just deal with
the 110F+ temps -- only 90+ overnight -- until I sort out what\'s going
on with it!]

 

[Don Y]

On 7/29/2020 3:49 PM, Joe Gwinn wrote:

And, what sort of potentials am I likely to see across them?

I have an old Branson.

Perhaps this is yours/of interest to you?

<http://www.marshallscientific.com/v//vspfiles/files/manuals/Branson450.pdf>

 

[Joe Gwinn]

On Fri, 31 Jul 2020 14:57:33 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:

On 7/29/2020 3:49 PM, Joe Gwinn wrote:
And, what sort of potentials am I likely to see across them?

I have an old Branson.

Perhaps this is yours/of interest to you?

http://www.marshallscientific.com/v//vspfiles/files/manuals/Branson450.pdf

What I have is a Branson 1200. Draws 0.7 amps at 120 Vac, or 82
watts, will deliver somewhat less in acoustic power. Got it in 2011.
I do have the schematic of that and related units, downloaded from the
Branson site before they were acquired. The drawings are dated circa
1990. No digital frippery.

Found my lab notes. Frequency is unstable, but around 50 KHz, pulsedd
at 120 Hz from the full-wave rectifier feeding the oscillator. The
peak voltage across the transducer is about 738 Vac peak to peak. Made
a simple voltage divider from two 1/2 watt carbon composition
resistors to protect the scope input.

The tank circuit is series resonant, a 1550 microhenry inductor
driving a 6.393 nanofarad PZT ceramic transducer with d=0.0031. This
with no water in the tank.

Joe Gwinn

 

[Don Y]

On 8/1/2020 10:05 AM, Joe Gwinn wrote:

On Fri, 31 Jul 2020 14:57:33 -0700, Don Y
blockedofcourse@foo.invalid> wrote:

On 7/29/2020 3:49 PM, Joe Gwinn wrote:
And, what sort of potentials am I likely to see across them?

I have an old Branson.

Perhaps this is yours/of interest to you?

http://www.marshallscientific.com/v//vspfiles/files/manuals/Branson450.pdf

What I have is a Branson 1200. Draws 0.7 amps at 120 Vac, or 82
watts, will deliver somewhat less in acoustic power. Got it in 2011.
I do have the schematic of that and related units, downloaded from the
Branson site before they were acquired. The drawings are dated circa
1990. No digital frippery.

From what I can see (user manual), my unit uses the smarts mainly as
a glorified timer <frown> There may be some crude PWM control for
the bath temperature (though I think it is all open loop)

Found my lab notes. Frequency is unstable, but around 50 KHz, pulsedd
at 120 Hz from the full-wave rectifier feeding the oscillator. The
peak voltage across the transducer is about 738 Vac peak to peak. Made
a simple voltage divider from two 1/2 watt carbon composition
resistors to protect the scope input.

The tank circuit is series resonant, a 1550 microhenry inductor
driving a 6.393 nanofarad PZT ceramic transducer with d=0.0031. This
with no water in the tank.

Thanks! I found some space in the garage (which means \"usable only
in the pre-dawn hours of the morning\") and started disassembling the unit.
(what a mess! SOME connections with spade lugs/connectors but others
hardwired. So, what value to the addition of those connectors?? <frown>)

Once sufficiently accessible that I can examine the HV board (the control
board is far too difficult to remove without rendering it unusable),
my first observation is a blown fuse on the mains (ceramic fuse so not
visually obvious).

This suggests Bill Beaty\'s comment re: shorted FETs may be the issue
(as mentioned, no real circuit protection from the mains -- other
than an EMI filter).

But, I\'ll have to do further disassembly in order to probe anything
further. (do folks make ANY effort to address serviceability?)

Only 364 more days of Summer remain so I should be able to get
to it, RSN! :-/

 

[Cydrome Leader]

Joe Gwinn <joegwinn@comcast.net> wrote:

On Thu, 30 Jul 2020 17:17:53 -0700, Don Y
blockedofcourse@foo.invalid> wrote:

On 7/30/2020 4:53 PM, Joe Gwinn wrote:
On Wed, 29 Jul 2020 17:30:51 -0700, Don Y
blockedofcourse@foo.invalid> wrote:

I was hoping to avoid probing the transducers directly. I figured I
could isolate the HV and just verify that the *drive* to it was
working properly. I.e., assume transformer and chokes are intact
so if driven correctly but still no apparent output power, the
transducers must be toast.

I don\'t think that not probing the transduced is a successful
approach. This is a power circuit - not too fussy. Just cob togther
a voltage divider.

I figure if the primary is being chopped, a lack of output power is
either:
- bad xformer
- bad chokes (all 3!)
- bad xducers (all 3)
In either case, I won\'t dick with it any further. (these sorts of things
turn up fairly regularly so its easier just to wait for another to come along)

The frequency will be 30 KHz to 50 KHz, most likely. Won\'t be MHz at
that power level, I\'d guess.

What is the make and model of your unit, anyway?

https://www.ebay.com/itm/VWR-Model-150D-Aquasonic-Ultrasonic-Bath-/143178745430

Apparently it has a different \"part number\" than \"model number\"
so chasing it down on the VWR web site was tedious.

WVR no longer offers this brand and model, it seems. But they ought
to have repair information. I\'d call them up on the phone and ask.
Also, see if they will tell you who made the unit for VWR.

That unit looks like a Crest, not sure how helpful they are with repairs
though.

Any photos of the \"mess\" inside?

 

[Joe Gwinn]

On Tue, 4 Aug 2020 07:40:47 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

On 8/1/2020 10:05 AM, Joe Gwinn wrote:

On Fri, 31 Jul 2020 14:57:33 -0700, Don Y
blockedofcourse@foo.invalid> wrote:

On 7/29/2020 3:49 PM, Joe Gwinn wrote:
And, what sort of potentials am I likely to see across them?

I have an old Branson.

Perhaps this is yours/of interest to you?

http://www.marshallscientific.com/v//vspfiles/files/manuals/Branson450.pdf

What I have is a Branson 1200. Draws 0.7 amps at 120 Vac, or 82
watts, will deliver somewhat less in acoustic power. Got it in 2011.
I do have the schematic of that and related units, downloaded from the
Branson site before they were acquired. The drawings are dated circa
1990. No digital frippery.

From what I can see (user manual), my unit uses the smarts mainly as
a glorified timer <frown> There may be some crude PWM control for
the bath temperature (though I think it is all open loop)

Found my lab notes. Frequency is unstable, but around 50 KHz, pulsedd
at 120 Hz from the full-wave rectifier feeding the oscillator. The
peak voltage across the transducer is about 738 Vac peak to peak. Made
a simple voltage divider from two 1/2 watt carbon composition
resistors to protect the scope input.

The divider is two half-watt carbon composition resistors, the first
being 1 Mohm, the second being 1 Mohm. The second has a 10 Mohm scope
probe across it. This dropped the voltage to about 60 Vpp.

The actual voltage division ratio was computed from the measured
resistance values, not the nominal values.

No bandwidth compensation capacitors were needed for such a low
frequency.

Half-watt carbon composition resistors are rated for 700 volts.

The tank circuit is series resonant, a 1550 microhenry inductor
driving a 6.393 nanofarad PZT ceramic transducer with d=0.0031. This
with no water in the tank.

Thanks! I found some space in the garage (which means \"usable only
in the pre-dawn hours of the morning\") and started disassembling the unit.
(what a mess! SOME connections with spade lugs/connectors but others
hardwired. So, what value to the addition of those connectors?? <frown>)

One can easily interrupt those paths without unsoldering.

Once sufficiently accessible that I can examine the HV board (the control
board is far too difficult to remove without rendering it unusable),
my first observation is a blown fuse on the mains (ceramic fuse so not
visually obvious).

This suggests Bill Beaty\'s comment re: shorted FETs may be the issue
(as mentioned, no real circuit protection from the mains -- other
than an EMI filter).

But, I\'ll have to do further disassembly in order to probe anything
further. (do folks make ANY effort to address serviceability?)

They are designed for servicing by module swapping by a harried tech,
not circuit tracing with scope.

VWR makes nothing, so they must have procured the units somewhere. I
looked at the Crest website, and they do look plausible. You may be
able to tell by the markings on the modules and boards.

..<https://www.crest-ultrasonics.com/>

Joe Gwinn

 

[Don Y]

On 8/6/2020 4:06 PM, Joe Gwinn wrote:

On Tue, 4 Aug 2020 07:40:47 -0700, Don Y <blockedofcourse@foo.invalid

Thanks! I found some space in the garage (which means \"usable only
in the pre-dawn hours of the morning\") and started disassembling the unit.
(what a mess! SOME connections with spade lugs/connectors but others
hardwired. So, what value to the addition of those connectors?? <frown>)

One can easily interrupt those paths without unsoldering.

Yes. So I can remove the line cord. And disconnect the transducers.
(We\'re talking about the HV board, here).

But, these conductors that go over to the \"control\" board are soldered
at the HV end and fitted to a multipin connector on the far end.

And, that connector also provides a means for other connections unrelated
to the HV board AND INSEPARABLE FROM THE UNIT ITSELF to mate to the control
board. (I suspect the heating element adhered to the side of the tank?)

So, the HV board can ALMOST be removed... but, not quite (I\'ll have to
remove those \"hardwired\" pins from the connector shell -- or physicaly
cut the conductors :< )

This suggests Bill Beaty\'s comment re: shorted FETs may be the issue
(as mentioned, no real circuit protection from the mains -- other
than an EMI filter).

But, I\'ll have to do further disassembly in order to probe anything
further. (do folks make ANY effort to address serviceability?)

They are designed for servicing by module swapping by a harried tech,
not circuit tracing with scope.

Yes. Though even that isn\'t easily supported with the situation
I outlined above. You can\'t swap out the HV module without:
- unsoldering those two connections at the HV board
- removing the control connector pins associated with them
- removing the control connector pins that tie into the \"other\" stuff
I.e., the manufacturer falls short of the \"module level FRU\" goal.

\"Made in Detroit\" <frown>

VWR makes nothing, so they must have procured the units somewhere. I
looked at the Crest website, and they do look plausible. You may be
able to tell by the markings on the modules and boards.

.<https://www.crest-ultrasonics.com/

I\'m encouraged (by the burned fuse) that the fault will not be subtle.
So, the downside is if there is propagation to other parts of the design
(esp the transducers) *or* if the fault was a cascade effect from
some problem elsewhere (e.g., the heating element, transducers, etc.)

Presently, making time to tinker with it is the problem... (already have
installed new showerheads so the need for this sort of \"toy\" isn\'t pressing).

I\'ll post back with the final resolution.

 

[Joe Gwinn]

On Thu, 6 Aug 2020 18:03:15 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

On 8/6/2020 4:06 PM, Joe Gwinn wrote:
On Tue, 4 Aug 2020 07:40:47 -0700, Don Y <blockedofcourse@foo.invalid

Thanks! I found some space in the garage (which means \"usable only
in the pre-dawn hours of the morning\") and started disassembling the unit.
(what a mess! SOME connections with spade lugs/connectors but others
hardwired. So, what value to the addition of those connectors?? <frown>)

One can easily interrupt those paths without unsoldering.

Yes. So I can remove the line cord. And disconnect the transducers.
(We\'re talking about the HV board, here).

But, these conductors that go over to the \"control\" board are soldered
at the HV end and fitted to a multipin connector on the far end.

And, that connector also provides a means for other connections unrelated
to the HV board AND INSEPARABLE FROM THE UNIT ITSELF to mate to the control
board. (I suspect the heating element adhered to the side of the tank?)

So, the HV board can ALMOST be removed... but, not quite (I\'ll have to
remove those \"hardwired\" pins from the connector shell -- or physicaly
cut the conductors :< )

It may be that 750 Vpp is too high for cheap-enough connectors, so
they decided to just live with unsoldering the HV cables. It should
be easy to uninstall the connector pins. There may be a special tool
for that, if the traditional piece of flat wire spring stock won\'t
work.

This suggests Bill Beaty\'s comment re: shorted FETs may be the issue
(as mentioned, no real circuit protection from the mains -- other
than an EMI filter).

But, I\'ll have to do further disassembly in order to probe anything
further. (do folks make ANY effort to address serviceability?)

They are designed for servicing by module swapping by a harried tech,
not circuit tracing with scope.

Yes. Though even that isn\'t easily supported with the situation
I outlined above. You can\'t swap out the HV module without:
- unsoldering those two connections at the HV board
- removing the control connector pins associated with them
- removing the control connector pins that tie into the \"other\" stuff
I.e., the manufacturer falls short of the \"module level FRU\" goal.

Check all the power-supply capacitors, not just the FETs.


>\"Made in Detroit\" <frown>

Yeah. Value Engineering, where they engineer all the value out.

VWR makes nothing, so they must have procured the units somewhere. I
looked at the Crest website, and they do look plausible. You may be
able to tell by the markings on the modules and boards.

.<https://www.crest-ultrasonics.com/

I\'m encouraged (by the burned fuse) that the fault will not be subtle.
So, the downside is if there is propagation to other parts of the design
(esp the transducers) *or* if the fault was a cascade effect from
some problem elsewhere (e.g., the heating element, transducers, etc.)

Presently, making time to tinker with it is the problem... (already have
installed new showerheads so the need for this sort of \"toy\" isn\'t pressing).

I\'ll post back with the final resolution.

OK. Don\'t forget to take pictures.

Joe Gwinn

 

[Don Y]

On 8/7/2020 10:16 AM, Joe Gwinn wrote:

On Thu, 6 Aug 2020 18:03:15 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

But, these conductors that go over to the \"control\" board are soldered
at the HV end and fitted to a multipin connector on the far end.

And, that connector also provides a means for other connections unrelated
to the HV board AND INSEPARABLE FROM THE UNIT ITSELF to mate to the control
board. (I suspect the heating element adhered to the side of the tank?)

So, the HV board can ALMOST be removed... but, not quite (I\'ll have to
remove those \"hardwired\" pins from the connector shell -- or physicaly
cut the conductors :< )

It may be that 750 Vpp is too high for cheap-enough connectors, so
they decided to just live with unsoldering the HV cables. It should
be easy to uninstall the connector pins. There may be a special tool
for that, if the traditional piece of flat wire spring stock won\'t
work.

I don\'t think this is the \"very high\" voltage that excites the
xducers. The two leads (black & white, #18AWG?) connect to
the HV board close to the bridge (which I suspect is directly
on the mains). As these are the only leads that interconnect
the HV board to the control board, I assume they control
the excitation.

No relay on the control board but perhaps a FET (?) hidden off in
a corner that\'s hard for me to observe (part number).

\"Made in Detroit\" <frown

Yeah. Value Engineering, where they engineer all the value out.

(sigh) If it was pieced together (instead of being designed as a unit),
then I can sort of see the reasoning behind the sloppy implementation.

OTOH, if this was the way it was INTENTIONALLY designed... <shakes head>

VWR makes nothing, so they must have procured the units somewhere. I
looked at the Crest website, and they do look plausible. You may be
able to tell by the markings on the modules and boards.

.<https://www.crest-ultrasonics.com/

I\'m encouraged (by the burned fuse) that the fault will not be subtle.
So, the downside is if there is propagation to other parts of the design
(esp the transducers) *or* if the fault was a cascade effect from
some problem elsewhere (e.g., the heating element, transducers, etc.)

Presently, making time to tinker with it is the problem... (already have
installed new showerheads so the need for this sort of \"toy\" isn\'t pressing).

I\'ll post back with the final resolution.

OK. Don\'t forget to take pictures.

<frown> I\'m lousy with close-in pix! Someday I should set up a
booth to better distribute the light (instead of relying on flashes).
And, having stuff \"in\" the enclosure just confuses the flash and focus!

 

[Joe Gwinn]

On Fri, 7 Aug 2020 14:07:38 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

On 8/7/2020 10:16 AM, Joe Gwinn wrote:
On Thu, 6 Aug 2020 18:03:15 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

But, these conductors that go over to the \"control\" board are soldered
at the HV end and fitted to a multipin connector on the far end.

And, that connector also provides a means for other connections unrelated
to the HV board AND INSEPARABLE FROM THE UNIT ITSELF to mate to the control
board. (I suspect the heating element adhered to the side of the tank?)

So, the HV board can ALMOST be removed... but, not quite (I\'ll have to
remove those \"hardwired\" pins from the connector shell -- or physicaly
cut the conductors :< )

It may be that 750 Vpp is too high for cheap-enough connectors, so
they decided to just live with unsoldering the HV cables. It should
be easy to uninstall the connector pins. There may be a special tool
for that, if the traditional piece of flat wire spring stock won\'t
work.

I don\'t think this is the \"very high\" voltage that excites the
xducers. The two leads (black & white, #18AWG?) connect to
the HV board close to the bridge (which I suspect is directly
on the mains). As these are the only leads that interconnect
the HV board to the control board, I assume they control
the excitation.

No relay on the control board but perhaps a FET (?) hidden off in
a corner that\'s hard for me to observe (part number).

\"Made in Detroit\" <frown

Yeah. Value Engineering, where they engineer all the value out.

(sigh) If it was pieced together (instead of being designed as a unit),
then I can sort of see the reasoning behind the sloppy implementation.

OTOH, if this was the way it was INTENTIONALLY designed... <shakes head

Modular designs are a compromise, trading flexibility for perfection.

VWR makes nothing, so they must have procured the units somewhere. I
looked at the Crest website, and they do look plausible. You may be
able to tell by the markings on the modules and boards.

.<https://www.crest-ultrasonics.com/

I\'m encouraged (by the burned fuse) that the fault will not be subtle.
So, the downside is if there is propagation to other parts of the design
(esp the transducers) *or* if the fault was a cascade effect from
some problem elsewhere (e.g., the heating element, transducers, etc.)

Presently, making time to tinker with it is the problem... (already have
installed new showerheads so the need for this sort of \"toy\" isn\'t pressing).

I\'ll post back with the final resolution.

OK. Don\'t forget to take pictures.

frown> I\'m lousy with close-in pix! Someday I should set up a
booth to better distribute the light (instead of relying on flashes).
And, having stuff \"in\" the enclosure just confuses the flash and focus!

You need a light tent, perhaps used outside, in the sunlight. I have
some LED studio lights for this.

What also works is a studio flash triggered by the camera flash.

To take pictures of equipment, need to disable face tracking et al.
Often, manual focus is best, or even simply necessary. If you have a
lot of light, the camera lens\' f-stop will be high, so the depth of
field will be great, for a reasonable shutter speed, so things become
far less critical. You cannot have too much light.

Joe Gwinn

 

[Don Y]

On 8/7/2020 3:17 PM, Joe Gwinn wrote:

On Fri, 7 Aug 2020 14:07:38 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

frown> I\'m lousy with close-in pix! Someday I should set up a
booth to better distribute the light (instead of relying on flashes).
And, having stuff \"in\" the enclosure just confuses the flash and focus!

You need a light tent, perhaps used outside, in the sunlight. I have
some LED studio lights for this.

I was going to build a little \"booth\" out of fabric and mount flood
lights behind the fabric (exterior to the enclosed space in the booth).
But, I need to find a way to make this reasonably large (as some of
the things I\'ve had to photograph are sizable) AND in such a way that
I can break it down and store it in a small space.

I had to do something similar with my green screen. These sorts of things
are inherently large but used so rarely that they don\'t justify the storage
space they would normally require!

What also works is a studio flash triggered by the camera flash.

To take pictures of equipment, need to disable face tracking et al.

I don\'t think any of my cameras track faces!

Often, manual focus is best, or even simply necessary. If you have a
lot of light, the camera lens\' f-stop will be high, so the depth of
field will be great, for a reasonable shutter speed, so things become
far less critical. You cannot have too much light.

The only cameras I have that give me that much control are film
cameras. (someday, I will by a new digital body).

[Actually, I\'ve an iPhone and a MAXX2 that *might* have fancier controls.
But, I\'ve never used them, so... <shrug> I\'m pretty sure none of the
other smart phones have more than nominal cameras.]

 

[Joe Gwinn]

On Fri, 7 Aug 2020 21:40:44 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

On 8/7/2020 3:17 PM, Joe Gwinn wrote:
On Fri, 7 Aug 2020 14:07:38 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

frown> I\'m lousy with close-in pix! Someday I should set up a
booth to better distribute the light (instead of relying on flashes).
And, having stuff \"in\" the enclosure just confuses the flash and focus!

You need a light tent, perhaps used outside, in the sunlight. I have
some LED studio lights for this.

I was going to build a little \"booth\" out of fabric and mount flood
lights behind the fabric (exterior to the enclosed space in the booth).
But, I need to find a way to make this reasonably large (as some of
the things I\'ve had to photograph are sizable) AND in such a way that
I can break it down and store it in a small space.

One can buy foldable light tents than can be put away when not in use.
They all come from China these days, and one might as well get Chinese
price as well. I\'ve had good luck with Neewer brand on Amazon.

<https://www.amazon.com/Neewer-Shooting-Diffusion-Backdrops-Photography/dp/B00GKGGICC/>

I had to do something similar with my green screen. These sorts of things
are inherently large but used so rarely that they don\'t justify the storage
space they would normally require!

Green screen?

What also works is a studio flash triggered by the camera flash.

To take pictures of equipment, need to disable face tracking et al.

I don\'t think any of my cameras track faces!

Often, manual focus is best, or even simply necessary. If you have a
lot of light, the camera lens\' f-stop will be high, so the depth of
field will be great, for a reasonable shutter speed, so things become
far less critical. You cannot have too much light.

The only cameras I have that give me that much control are film
cameras. (someday, I will by a new digital body).

[Actually, I\'ve an iPhone and a MAXX2 that *might* have fancier controls.
But, I\'ve never used them, so... <shrug> I\'m pretty sure none of the
other smart phones have more than nominal cameras.]

One can get cradles for cell phones, so they can be mounted on a
tripod. The other trick is to use the self timer in the phone, so the
photo is taken ten or fifteen seconds after the button is pushed.

With enough light, everything will be in focus, not just faces.

The modern web catalog photo scheme is a LED ringlight with a
cellphone in cradle mounted in the center of the ring, all on a tripod
stand of some kind. Great for beauty shots, if the ads are any
indication.

Here is a representative full kit; there are many options:

<https://www.amazon.com/Neewer-Center-Carrying-YouTube-Shooting/dp/B01JIBWCX4/>

People also use a digital SLR with a light tent.


Joe Gwinn

 

[Don Y]

On 8/9/2020 11:22 AM, Joe Gwinn wrote:

On Fri, 7 Aug 2020 21:40:44 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

You need a light tent, perhaps used outside, in the sunlight. I have
some LED studio lights for this.

I was going to build a little \"booth\" out of fabric and mount flood
lights behind the fabric (exterior to the enclosed space in the booth).
But, I need to find a way to make this reasonably large (as some of
the things I\'ve had to photograph are sizable) AND in such a way that
I can break it down and store it in a small space.

One can buy foldable light tents than can be put away when not in use.
They all come from China these days, and one might as well get Chinese
price as well. I\'ve had good luck with Neewer brand on Amazon.

Yes, but they all are small-ish. I want to be able to do SfM photogrammetry
*inside* the \"tent\", for reasonably large objects. So, I need the (motorized)
camera positioning mechanism to be able to fit inside the tent. The mechanism,
of course, needs to surround the subject. So, for even small subjects, you
need a big volume of \"uniform light\".

This lets me build 3D models to import into other design documents, etc.

I\'d hoped to be fabricating that mechanism -- along with a couple of
panoheads -- this Fall/Winter, but so many things are still shut down
that I\'ve had to rethink that schedule (no problem as \"stuff\" always
finds a way to fill the scheduling space emptied by such events!)

[I can work-around a lack of panhead but, if I\'m building camera
fixtures, building yet another is relatively simple]

I had to do something similar with my green screen. These sorts of things
are inherently large but used so rarely that they don\'t justify the storage
space they would normally require!

Green screen?

It\'s a (common) technique used for chroma key compositing. You see it
every night on the weather report. In real life, the weatherperson stands
in front of a homogeneous (green) backdrop -- a green screen -- and looks
off camera to see a live image of his/herself superimposed on a computer
animation (or other video source). The TV viewer sees this composite
image, as well. So, when the weatherperson APPEARS to be pointing at
something IN the animation, the broadcast signal actually shows that.

I have hardware (& software) that lets me superimpose live video over
interactive (as well as canned) computer-generated visualizations.
To simplify isolating the \"foreground\" (i.e., the weatherperson) from
the computer generated video, the subject is filmed standing in front
of that \"green screen\".

(other colors are possible; green is for hysterical raisins -- consistent
with NOT conflicting with the garments that weatherpeople tend to wear)

Hazards: <https://www.youtube.com/watch?v=PUkZsfqPmX8>

Often, manual focus is best, or even simply necessary. If you have a
lot of light, the camera lens\' f-stop will be high, so the depth of
field will be great, for a reasonable shutter speed, so things become
far less critical. You cannot have too much light.

The only cameras I have that give me that much control are film
cameras. (someday, I will by a new digital body).

[Actually, I\'ve an iPhone and a MAXX2 that *might* have fancier controls.
But, I\'ve never used them, so... <shrug> I\'m pretty sure none of the
other smart phones have more than nominal cameras.]

One can get cradles for cell phones, so they can be mounted on a
tripod. The other trick is to use the self timer in the phone, so the
photo is taken ten or fifteen seconds after the button is pushed.

My solution will be a digital body for my SLR -- reuse all of the
kit that I\'d previously accumulated. But, putting real FILM in
it and scanning the photos after they are developed is just not
a practical way to \"snap a few photos\"! :>

I now use a phone primarily to capture things that I see while driving
or large pieces of equipment -- where the important feature is the
size/mass of the item and not \"the clarity of the logo on the face\".
I can keep a phone in the car whereas I\'d not waste the space
that a \"real\" camera required.

[it seems like cars have less out-of-sight storage space than ever
before! I can barely get the spare tire in the trunk on her car.
And, the glove box is barely big enough to store a sandwich (not that
I\'d store a sandwich, there!)]

With enough light, everything will be in focus, not just faces.

The modern web catalog photo scheme is a LED ringlight with a
cellphone in cradle mounted in the center of the ring, all on a tripod
stand of some kind. Great for beauty shots, if the ads are any
indication.

Yes. I have all that kit -- assorted lenses, ring light, flash units,
stands for extreme close-up shots (e.g., jewelry) -- but for FILM. My
interest in photography is many decades out of date. :<

Hence the digital camera body is the most expeditious solution.

Here is a representative full kit; there are many options:

https://www.amazon.com/Neewer-Center-Carrying-YouTube-Shooting/dp/B01JIBWCX4/

People also use a digital SLR with a light tent.

 

[Joe Gwinn]

On Sun, 9 Aug 2020 13:19:26 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

On 8/9/2020 11:22 AM, Joe Gwinn wrote:
On Fri, 7 Aug 2020 21:40:44 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

You need a light tent, perhaps used outside, in the sunlight. I have
some LED studio lights for this.

I was going to build a little \"booth\" out of fabric and mount flood
lights behind the fabric (exterior to the enclosed space in the booth).
But, I need to find a way to make this reasonably large (as some of
the things I\'ve had to photograph are sizable) AND in such a way that
I can break it down and store it in a small space.

One can buy foldable light tents than can be put away when not in use.
They all come from China these days, and one might as well get Chinese
price as well. I\'ve had good luck with Neewer brand on Amazon.

Yes, but they all are small-ish. I want to be able to do SfM photogrammetry
*inside* the \"tent\", for reasonably large objects. So, I need the (motorized)
camera positioning mechanism to be able to fit inside the tent. The mechanism,
of course, needs to surround the subject. So, for even small subjects, you
need a big volume of \"uniform light\".

They do come bigger, I have a 48\" cube one from someone else, but
people use softboxes for that size and above.

This lets me build 3D models to import into other design documents, etc.

I\'d hoped to be fabricating that mechanism -- along with a couple of
panheads -- this Fall/Winter, but so many things are still shut down
that I\'ve had to rethink that schedule (no problem as \"stuff\" always
finds a way to fill the scheduling space emptied by such events!)

[I can work-around a lack of panhead but, if I\'m building camera
fixtures, building yet another is relatively simple]

I had to do something similar with my green screen. These sorts of things
are inherently large but used so rarely that they don\'t justify the storage
space they would normally require!

Green screen?

It\'s a (common) technique used for chroma key compositing. You see it
every night on the weather report. In real life, the weatherperson stands
in front of a homogeneous (green) backdrop -- a green screen -- and looks
off camera to see a live image of his/herself superimposed on a computer
animation (or other video source). The TV viewer sees this composite
image, as well. So, when the weatherperson APPEARS to be pointing at
something IN the animation, the broadcast signal actually shows that.

Ahh. I knew them as blue screens - it was funny when the newscaster
was transparent where his tie or shirt was. Probably why they went to
green.

I have hardware (& software) that lets me superimpose live video over
interactive (as well as canned) computer-generated visualizations.
To simplify isolating the \"foreground\" (i.e., the weatherperson) from
the computer generated video, the subject is filmed standing in front
of that \"green screen\".

(other colors are possible; green is for hysterical raisins -- consistent
with NOT conflicting with the garments that weatherpeople tend to wear)

Hazards: <https://www.youtube.com/watch?v=PUkZsfqPmX8

Yep. Blue!

Often, manual focus is best, or even simply necessary. If you have a
lot of light, the camera lens\' f-stop will be high, so the depth of
field will be great, for a reasonable shutter speed, so things become
far less critical. You cannot have too much light.

The only cameras I have that give me that much control are film
cameras. (someday, I will by a new digital body).

[Actually, I\'ve an iPhone and a MAXX2 that *might* have fancier controls.
But, I\'ve never used them, so... <shrug> I\'m pretty sure none of the
other smart phones have more than nominal cameras.]

One can get cradles for cell phones, so they can be mounted on a
tripod. The other trick is to use the self timer in the phone, so the
photo is taken ten or fifteen seconds after the button is pushed.

My solution will be a digital body for my SLR -- reuse all of the
kit that I\'d previously accumulated. But, putting real FILM in
it and scanning the photos after they are developed is just not
a practical way to \"snap a few photos\"! :

I now use a phone primarily to capture things that I see while driving
or large pieces of equipment -- where the important feature is the
size/mass of the item and not \"the clarity of the logo on the face\".
I can keep a phone in the car whereas I\'d not waste the space
that a \"real\" camera required.

[it seems like cars have less out-of-sight storage space than ever
before! I can barely get the spare tire in the trunk on her car.
And, the glove box is barely big enough to store a sandwich (not that
I\'d store a sandwich, there!)]

With enough light, everything will be in focus, not just faces.

The modern web catalog photo scheme is a LED ringlight with a
cellphone in cradle mounted in the center of the ring, all on a tripod
stand of some kind. Great for beauty shots, if the ads are any
indication.

Yes. I have all that kit -- assorted lenses, ring light, flash units,
stands for extreme close-up shots (e.g., jewelry) -- but for FILM. My
interest in photography is many decades out of date. :

I have all that legacy kit too. Olympus OM1 and successors. (Nikon
too big and heavy, but damn rugged.) A lot of it can be and was
reused. I do have my old film cameras, but for historical interest
only. I\'m all digital now. Although it\'s still useful to know what a
f-stop is.

What killed film for me, long before digital was clearly better, was
that it became impossible to get adequate developing, even for
Kodacolor (which can be done at home), never mind Kodachrome (not a
prayer).


>Hence the digital camera body is the most expeditious solution.

I did that for a while, but no longer schlep all that stuff when
travelling. Now, it\'s a Canon PowerShot G7X Mark II. But the
cellphone cameras are snapping at the heels of traditional cameras.
The CMOS sensors are way better than film ever was, even in
consumer-grade cameras.

The PS G7 Mk2 does have a manual focus mode, and one can disable the
face tracking. This works well for technical photography.

Joe Gwinn

Here is a representative full kit; there are many options:

https://www.amazon.com/Neewer-Center-Carrying-YouTube-Shooting/dp/B01JIBWCX4/

People also use a digital SLR with a light tent.

 

[Don Y]

On 8/11/2020 5:44 PM, Joe Gwinn wrote:

On Sun, 9 Aug 2020 13:19:26 -0700, Don Y <blockedofcourse@foo.invalid
wrote:

One can buy foldable light tents than can be put away when not in use.
They all come from China these days, and one might as well get Chinese
price as well. I\'ve had good luck with Neewer brand on Amazon.

Yes, but they all are small-ish. I want to be able to do SfM photogrammetry
*inside* the \"tent\", for reasonably large objects. So, I need the (motorized)
camera positioning mechanism to be able to fit inside the tent. The mechanism,
of course, needs to surround the subject. So, for even small subjects, you
need a big volume of \"uniform light\".

They do come bigger, I have a 48\" cube one from someone else, but
people use softboxes for that size and above.

I was thinking of using the \"collapsible\" (poor choice of words)
screens that one places behind the windshield to shade the vehicle
interior. These fold in on themselves in slick motions to occupy
a fraction of their expanded size. As the fabric is just serving as
a diffuser, it need not be very robust.

Green screen?

It\'s a (common) technique used for chroma key compositing. You see it
every night on the weather report. In real life, the weatherperson stands
in front of a homogeneous (green) backdrop -- a green screen -- and looks
off camera to see a live image of his/herself superimposed on a computer
animation (or other video source). The TV viewer sees this composite
image, as well. So, when the weatherperson APPEARS to be pointing at
something IN the animation, the broadcast signal actually shows that.

Ahh. I knew them as blue screens - it was funny when the newscaster
was transparent where his tie or shirt was. Probably why they went to
green.

Yes. The choice of color depends on what you\'re NOT likely to encounter
in the \"precious\" part of the video. In theory, you can use any (solid)
color (uniformly lit!). You\'d not want to choose a \"flesh tone\"!

Movies tend to use green and blue. Most live broadcasts (weather)
that I\'ve seen use green.

Note that you can use this effect in post, as well, as many of the
non-linear editing tools let you \"pick\" the color that will be
masked out (using an eyedropper tool to sample the actual video).

My hardware/software tools let you interact with the \"background\"
as it is happening -- instead of trying to synchronize two independant
video streams (one being masked) after-the-fact.

[Note most weather/sports folks *control* the animation so they can
interact with it properly]

With enough light, everything will be in focus, not just faces.

The modern web catalog photo scheme is a LED ringlight with a
cellphone in cradle mounted in the center of the ring, all on a tripod
stand of some kind. Great for beauty shots, if the ads are any
indication.

Yes. I have all that kit -- assorted lenses, ring light, flash units,
stands for extreme close-up shots (e.g., jewelry) -- but for FILM. My
interest in photography is many decades out of date. :

I have all that legacy kit too. Olympus OM1 and successors. (Nikon
too big and heavy, but damn rugged.) A lot of it can be and was
reused. I do have my old film cameras, but for historical interest
only. I\'m all digital now. Although it\'s still useful to know what a
f-stop is.

I haven\'t snapped a film photo in decade(s). But, the kit doesn\'t
\"need watering\" and is quite content to reside on a closet shelf.
So, the only advantage to discarding it would be if I needed to
reclaim that shelf! (I\'d rather discard some suits or other
articles of clothing if space was that precious! :> )

What killed film for me, long before digital was clearly better, was
that it became impossible to get adequate developing, even for
Kodacolor (which can be done at home), never mind Kodachrome (not a
prayer).

I never bothered trying to develop film. As long as I could get prints
in some \"reasonable\" amount of time, it wasn\'t an issue.

What drew me to the digital alternative was the \"polaroid\" immediacy
of it and the near-infinite number of pix that I could take on a
\"roll\".

[My other half has learned the downside of taking too many photos
as she spends hours trying to catalog them!]

Hence the digital camera body is the most expeditious solution.

I did that for a while, but no longer schlep all that stuff when
travelling.

I don\'t \"sightsee\" anymore. When I travel, it\'s almost always for an
off-site get-together with colleagues. I have no desire to collect
pictures of that motley crew! <grin> And, anything that I might want
to capture -- that I can\'t just have them email me a copy of -- I can
get with a cell phone camera.

In recent memory, I\'ve snapped photos of the (large) wildfire
\"up the street\". And, heavier rainfalls (when you see a SHEET of
water come off the roof it is impressive; or, 8 inches of standing
water in the yard!). Plus things that are hard to describe, to
someone who\'s never carefully examined such (e.g., my electric
wheelbarrow). Or, images of our citrus harvests (several hundred
pounds of oranges are kind of hard to visualize). Or, \"wiring or
assembly details\" so I don\'t have to take detailed notes!

It\'s never been driven by artistry or great detail!

Now, it\'s a Canon PowerShot G7X Mark II. But the
cellphone cameras are snapping at the heels of traditional cameras.
The CMOS sensors are way better than film ever was, even in
consumer-grade cameras.

I suspect I could probably dig a better (cell phone) camera out
of my box of rescues. There\'s an iPhone 6S in the car and a
Moto MAXX 2 sitting on a shelf in my bedroom. But, I much prefer
this *small* LG as I can slip it in a pocket without feeling
like it\'s going to snap in half due to its size/length!

The PS G7 Mk2 does have a manual focus mode, and one can disable the
face tracking. This works well for technical photography.

I only snap \"formal\" photos just prior to publishing documents.
At that time, I can organize everything that I need to photograph
and \"stage\" it properly. Otherwise, the \"overhead\" is too costly.