Another simple Passive IR clock control alternative...

250 views
Skip to first unread message

Robert L

unread,
Sep 22, 2017, 1:10:27 PM9/22/17
to neonixie-l
Hi folks,

Pretty much all my clocks have some form of passive IR control to avoid burning tubes when no one is looking... In many cases the PIR control is accomplished with modifications to the controller board, but some of the clocks are more of a pain to modify than others. In particular, on one of my small VFD clocks it's difficult to modify the base board to remove filament voltage and I'm tired of replacing failed VFD tubes on this unit.

I just came across low cost PIR power switches used to control the DC supplied to LED strips and similar lights. The switch I used is rated for 5 - 24 VDC at 5 amps. It's available with fast delivery time from Amazon and for a lot lower cost on ebay, but with very long delivery time. Here's the Amazon part for the impatient among us:


I'm using the PIR switch in three configurations at the moment...
  • An unmodified PIR switch in series with the 5V wall wart powering an old USB hub. I added a jumper internal to the hub to supply power to the USB ports with no computer connection. I also wired the red hub LED to light when power is applied. The stock coaxial connectors matched the hub connectors.


  • An unmodified PIR switch in series with the 12V wall wart powering a clock. As with the USB hub above, the stock coaxial connectors happened to be the proper size.
  • I modified the wired connections to a PIR switch so that it's in series with power lines on a USB extension cable. The data lines are disconnected and the extension cable powers a clock running off a plug-in 5V USB power cube.





Note that I also modified the PR module by changing a cap to extend the power-on time when motion is detected. Stock time is adjustable from a few seconds up to about 6 minutes. As modified, the on-time ranges from 2 to 39 minutes.

Change the circled cap to a 0.1 uF/0603 part for the longer on-time.

Enjoy...

Bob

Paolo Cravero

unread,
Sep 30, 2017, 6:18:47 AM9/30/17
to neoni...@googlegroups.com
That's interesting Robert, thank you for sharing. I am not so comfortable and fast with case building, so the less holes I have to drill, the better ;-)
 

... on one of my small VFD clocks it's difficult to modify the base board to remove filament voltage and I'm tired of replacing failed VFD tubes on this unit.


AFAIK, VFD tubes are very long lasting as long as their filaments are treated correctly. What kind of VFDs are experiencing such a high failure rate?


 
I just came across low cost PIR power switches used to control the DC supplied to LED strips and similar lights.

I will look for a similar device that ships to Italy since your Amazon link seller does not. Well, on the big Chinese marketplace they go for 4 €/$ each.

Paolo

gregebert

unread,
Sep 30, 2017, 11:57:14 AM9/30/17
to neonixie-l
The VFD for the display on my kitchen's oven runs 24/7, and after 12 years there are very noticeable dark-bands near the filament, which is probably the result of higher bombardment of electrons onto the phosphor. The filament does not glow bright enough to see it, and it's always on.

So, if you have VFD's I guess you want to keep the filaments always-on, and blank the anodes when not in use.
Message has been deleted

Robert L

unread,
Sep 30, 2017, 7:23:48 PM9/30/17
to neonixie-l
I've also had generally very good reliability with vfd displays... instrument displays, larger tubes in clocks... 

Tubes are in this clock are Russian IV-6 with roughly 1.0 to 1.1 V across the filament. Voltage is limited by individual series resistors from the nominal 5V supply. Filaments are dim, but visible in a less than bright room and vary in intensity. See photos below.

As Greg suggests, "So, if you have VFD's I guess you want to keep the filaments always-on, and blank the anodes when not in use." That's exactly what I originally did with this clock.


As a side note, it would be easy to replace the (crude but reasonably effective!) series current limit resistors with active high side transistor current limiters... but all tubes are within spec as I understand it. 

B

gregebert

unread,
Sep 30, 2017, 11:42:31 PM9/30/17
to neonixie-l
I did some tradeoff analysis on filament drivers for a project underway, and I concluded a dropping resistor is better than a current-regulator for filaments, at least in my case.

To cancel-out the voltage-gradient along a filament, AC rather than DC is commonly used. That makes a current-regulator more complex even if you just use a square-wave.

The basic problem with filaments is that their cold-resistance is several times lower than their operating resistance, and without any current-limiting they will undergo a surge-current at turn-on that is many times greater than their operating current. It's basically why incandescent bulbs almost always fail when switched-on.

If you drive a 1V filament from a constant 1V supply, the filament current is entirely determined by the filament's resistance. When it's energized, it rapidly heats up, which increases the resistance which lowers the current, hence lowers the temperature. The current decreases until equilibrium is reached.

The filament in the tube I'm working with has a cold-resistance of 2.7 ohms, and measured 7 ohms at it's operating current of 200mA. That's almost a 3:1 range.

If you drove this from a 5V supply, you would need about 18ohms of series resistance for 200mA at 5V. When you first turn it on, the surge current will be 5/(2.7+18) = 241mA. On the other hand, if you just used the 1V supply, the surge-current is 370mA.  Clearly the higher-voltage supply with the dropping resistor will put far less stress on the tube. The tradeoff is wasted energy (280mW for the tube, vs 720mW for the dropping resistor).

I did some web research on this, and found a good paper about extending the life of radio-transmitter tubes. They basically do the same thing: Add series-resistance to limit the surge current, and the tube life was dramatically increased. I think the same reasoning can apply to VFDs, magic eyes, & NIMOs.

The one thing I could not draw a conclusion on was how to determine when to leave filaments running, vs turning them off. Even with current-limiting a filament will undergo mechanical degradation from expansion and contraction when power-cycled. But a filament will also undergo some degradation just from being hot. I wont have enough time and tubes to determine this thru experiments. I've seen some info about thoriated filaments (ie, containing Thorium) degrading if they are kept idle at currents significantly below their normal operating current. I'm inclined to keep my clock filaments on while the clock is plugged-in, energize them via a PIR sensor, and unplug the clock when I put it away for a few months. 

Robert L

unread,
Oct 1, 2017, 2:40:37 AM10/1/17
to neonixie-l
Hey Greg,

Filament voltage is only 1 V on these IV-6 tubes, so not a lot of gradient to speak of... and the active current limiter will take care of any in-rush current. Still, I'm seeing relatively well matched voltage drop across the 6 filaments so the simple current limit resistors already in place seem to be doing an adequate job of controlling the operating current.

I'll see what happens over time with the switched power on these IV-6 tubes... The good thing is it's just one clock and the tubes are cheap!

I've another clock with cold filaments when in PIR sleep... that clock uses a series connection of three tube filaments to limit the filament voltage - there's no series resistor. I'll also keep an eye on that one as it faces a double whammy - inrush current and filaments switched off during sleep!

Regards,
Bob

Quixotic Nixotic

unread,
Oct 1, 2017, 2:06:48 PM10/1/17
to neoni...@googlegroups.com
A bit late of me, I apologise it was a week or so ago, but for those of you that did not manage to visit Paul Parry in London here are a few pictures of the Bad Dog Designs stand at the Olympia exhibition hall. Paul, it was great to meet you and see how enthusiastic both you and the general public were. Dalibor's tubes were simply a work of art to see and I got to hold one in my hand - they are really something very special. The sound-to-light dekatron eyebrows on the robot were most amusing - I felt we established an interpersonal rapport of kinds - may he rust in peace, if not in pieces.

IMG_1163 copy.jpg
IMG_1157 copy.jpg
Message has been deleted

Robert L

unread,
Oct 2, 2017, 10:02:40 PM10/2/17
to neonixie-l
So I considered using a Negative Temp Coefficient (NTC) inrush limiter for my clock with the three IV-11 filaments in series on the 5V rail... Found an appropriate part after a pretty broad search, but ran into a 100 piece minimum order quantity at $2.50 per each and 6-8 weeks lead time. No... I don't think so... not a good plan...

Here's plan "B" which I actually like a lot more than plan "A" above. 

The design below uses a mosfet with turn-on time slowed by an RC network. As shown, the slow start takes about a quarter second to rise to near 5V. I believe this will be plenty slow enough to protect the IV-11 filaments. I've previously used the same basic circuit to allow a switch mode wall wart to power a device with high initial inrush charging capacitors. The wall wart would go into hiccup mode trying to start the device in question when used without the inrush limiter. This slow start is a broadly useful circuit.

(Here's a TI note describing the circuit below as well as several others:  http://www.ti.com/lit/an/slva156/slva156.pdf)

A few nice things about this circuit... 

1) It will work to slow start most any 5 - 12 V load (though possibly with component value changes).
2) Operating voltage drop can be made very small with careful mosfet selection. My IV-11 will actually like a slightly lower operating voltage as the are currently at the upper end of spec. 
3) Reset time is on the order of 1.5 seconds as shown. (Reset time is the time after power off that is needed to assure a slow start the next time power is applied. This time allows a large cap in the circuit to discharge. This is rather like the time needed for an NTC inrush protector to cool between applications of power.)
4) The circuit can be set up to run quite cool depending on operating load and selection of a mosfet with sufficiently low Ron.
5) One slow start circuit will take care of multiple loads on a common power supply... this was not going to work with the particular NTC part I had found due to excessive voltage drop if I used just one part for multiple filament strings. This one circuit can actually replace a wide variety of NTC parts for these low voltage applications.

So... I'll spin up a small pcb for the slow start to be included in unused area on the next larger board that I have occasion to build. I expect there will be other uses for the little board so worth the trouble to make a small pcb for it. Will also fine tune a few values as needed but the values shown here are a not unreasonable starting place.



Onward...
Bob

Robert L

unread,
Oct 2, 2017, 10:12:34 PM10/2/17
to neonixie-l
For the circuit shown and simulated above... there's output rise time and output delay from application of power. These are two different times.

In the simulation below, + 5 input power comes on at T=0 and it then takes about a half second for the slow start output to reach ~5 V.  The slow start voltage rises from 0 to ~5V in about 240 mS... This is the ramp the load sees. There's a gap after the application of power before the output begins to rise.

B

Terry S

unread,
Oct 2, 2017, 11:16:53 PM10/2/17
to neonixie-l


On Monday, October 2, 2017 at 9:02:40 PM UTC-5, Robert L wrote:
So I considered using a Negative Temp Coefficient (NTC) inrush limiter for my clock with the three IV-11 filaments in series on the 5V rail... Found an appropriate part after a pretty broad search, but ran into a 100 piece minimum order quantity at $2.50 per each and 6-8 weeks lead time. No... I don't think so... not a good plan...


Did you consider asking for samples? Most manufacturers are happy to provide them.
Terry

Robert L

unread,
Oct 3, 2017, 12:58:37 AM10/3/17
to neonixie-l
Hi Terry,

I requested samples, but no joy... Fast and courteous response from the factory, but no parts available.

Bottom line, I like the increased flexibility of the mosfet design better... work in progress / sketch below... should be able to shrink it a bit more.



GastonP

unread,
Oct 3, 2017, 9:27:09 AM10/3/17
to neonixie-l
Pleasae be careful with the operating limits of the MOSFET... the part you are using has a Gate-Source maximum voltage (Vgss) of 8 volts and in spite of being zener protected, you should not routinely use that protection as a design feature.
I'd say that this part is safe for this circuit when the input voltages are lower than 7 volts. Other MOSFETS have higher Vgss that can be used to your advantage. I.e. the BSS84 is also a PMOS but with a Vgss of 20 Volts.

Gastón

On Monday, October 2, 2017 at 11:02:40 PM UTC-3, Robert L wrote:

A few nice things about this circuit... 

1) It will work to slow start most any 5 - 12 V load (though possibly with component value changes).
 



Onward...
Bob

Robert L

unread,
Oct 3, 2017, 11:58:23 AM10/3/17
to neonixie-l
Hi Gaston,

Good points... Going deeper on where I am with the design at the moment...

I considered the BSS84 early on as I have some in the spare parts bin... but...
  > BSS84 is a nice part and I use it a lot, but only rated for 130 mA and I'll be running around 200 mA continuous in this application. Also has significant Rds(on) at -5 V of 10 ohms max... can be too high for my application.
  > Si1013R is rated 350 mA continuous at 25C and Rds(on) at 4.5 V of 1.2 ohms max. Simulation already done to confirm Vgs comfortably within limits as circuit operates and transitions on->off and off-> on.
So getting closer with the Si1013R (which also happened to be in the built-in LTSpice library). But, as you point out, the Si1013R will be in trouble with a 12V rail... so I'm still looking.

I'd really like to build one little board for a wide range of applications... at least 12V operating, at least 1 A operating.

Currently leaning toward the NTF5P03T3G... will probably stop looking as this is a more than adequate selection for current and anticipated needs. SOT-223 tabbed package that is still easy for hand solder if necessary, way lower Rds(on), ample voltage and current limits. Alternative parts available in the same package and pin-out...

Thought about a dual footprint board so that I can load a SOT-23 if I want for a given application... will probably just use the larger 223 and be done with it. Typically have the modest additional space needed.... say 1" x 0.4 " or a bit smaller total board size with the 223.



I'll update the post when I have parts selected and a layout I like.

B
Message has been deleted

Robert L

unread,
Oct 4, 2017, 6:15:14 AM10/4/17
to neonixie-l
Dual footprint for SOT-23 and SOT-223... Heat mass moved more to center of the small pcb.


Robert L

unread,
Oct 4, 2017, 8:41:34 PM10/4/17
to neonixie-l
Looking good to me in sim land... circuit meets my design objectives up to and beyond 12 VDC with loads above 3A. 

Caution: Watch out for heat with higher loads... Small pcb shown will not support the resulting dissipation with these high loads...





Will add this to the next board I turn where there's a bit of extra space.

I'm curious... If I were to make extra of these now would others want a bare board and simple parts kit? Bare fab, transistor and the four 0805 passives only. Let me know if you're interested? I'd make a run of boards specifically for these now rather than waiting for another excuse down the road.

Figure on the order of $3.00 to $4.00/ea delivered in an envelope by first class mail. Cost will depend on how many ways we share the board cost which dominates price. 

No commitment on my part as yet, just seeing if there's interest...

Regards,
Bob

gregebert

unread,
Oct 4, 2017, 11:21:05 PM10/4/17
to neonixie-l
Before you fab it, you might want to put some protection between gate & source to prevent ESD damage since this is a separate PCB and wont have the benefit of surrounding circuity until after it's connected. A large resistor (10meg) across C1 will keep C1 discharged. It's not entirely obvious, but C1 is actually a good form of ESD protection, because it can absorb a fairly decent amount of energy before Vgs rises above a safe level.

The human body model (HBM) for ESD testing is 100pF. If I did my math correctly, a 2kV zap at the input terminals would only produce about 100mV across the gate-source assuming C1 has low inductance and low ESR

For the output terminal, there's a parasitic diode in the MOSFET to protect from positive zaps. To protect from negative zaps at the output, I think a reverse-biased diode from Vload to GND would work.
Message has been deleted

Robert L

unread,
Oct 5, 2017, 12:28:24 PM10/5/17
to neonixie-l
Thanks Greg... Good points... 

The circuit is intended to be installed internal to a housed device, but ESD definitely an issue prior to and during installation. Adding belt and suspenders can't hurt.

 Now 1.0" x 0.4"


Additional comments and suggestions always welcome...


B



Alex

unread,
Oct 7, 2017, 6:40:33 AM10/7/17
to neonixie-l
Comments from a PCB layout perspective would only really be focused on the size of your output trace (as an outside casual not done the maths comment) - bearing in mind the 1-3A rating you mention and seeing as it passes through 0805 parts. Only really thinking that it would probably survive a brief output short otherwise... Also heat dissipation from the FET, I can see Gnd going to the green pour which I assume is the bottom copper but not much else? I would normally have pepper vias all round the FET, with as much copper pegged to it as possible on both sides... 

What layout package is that by the way?

- Alex

Robert L

unread,
Oct 7, 2017, 12:33:27 PM10/7/17
to neonixie-l
Hi Alex,

Thanks for the comments!

Circuit includes a comment that heat will be an issue at higher loads. Components will go 12 V or more at 3 A with small value changes to adjust turn on ramp, but highest I imagine for this implementation will be 12V at 1A which was my design point.

Looking at 6 mW in the immediate application (5V, 200 mA),  80 mW for the 12V, 1A case. Traces are adequate at 0.020" with a short neck down to 0.015 through the 0805 parts. Note also low Rds(on) of 150 milliohm for this MOSFET at a worst corner case.

Layout is ExpressPCB as I have several existing small utility boards already in ExpressPCB. I'll probably make a few more of the other small cards when I do this one so leaving the board in ExpressPCB.

I'll toss in a few heat spreading via, but not going to do much in this case. I frequently end up with additional layers and multiple via when building designs that operate at higher dissipation - just not the case here! 

B
Reply all
Reply to author
Forward
0 new messages