Charging caps via PV without a circuit?

[abit]

Greetings,

I signed up because I have a simple question I cannot seem to get my brains around

If I have a solar panel with a 6V and 16mA open values.. if I connect a 10u ceramic with a 1N4148 diode in series like [solar] -> [diode] -> [cap] and close the circuit with the solar 'ground'...

The solar panel should eventually charge the cap to full 6V-0.200V or so, right? As the cap gets charged, the ESR in both the PV and the CAP decrease, so the PV pumps more and more voltage through the diode until it balances to a point where the diode reverse leakage and the cap's internal leakage balance?

That is, efficiency aside - the cap is going to charge without any additional circuits to as near to the PV's capacity as time and the PV's maximum output allows it?

For example, if I do this in a shade, the PV's open voltage is going to be limited to say 20%, or 2V.. the diode drops another 0.2V or so, and the cap is at 1.8V - before the sun comes out. Then the PV swings to full capacity, and again charges the cap to a 6V maximum minus the 0.2V diode drop which keeps trickling in some uA to replenish the cap's internal leakeage..

These are so small values I cannot measure with a meter, so I need to just think it through to understand... Is this about right?

This is kinda important because the caps are input to an IC with 5.5V absolute max, and the panels are rated just above that. Thanks!


Updated:

Thank you all so much!
I now have all the parts I need at hand, and I started soldering a 1000uF bank made out of 100 pieces of 10u 10V X5R 1210 caps in parallel. I have some fixtures to line up the caps, so this is not as crazy or time-consuming as it sounds. Maybe a few years from now 100u of the same package is going to be everywhere to make 10,000uF banks, and the banks may further parallel...

It's use is in an LED-lit magnifier, which is turned ON/OFF by a push-button for only 10-30 seconds at a time.. Then it sits idle for 10 or more minutes, solar-side up under daylight or artificial light to slowly re-charge the caps.. Again, efficiency is not an issue, not even cost really at this point. And I do not yet have an idea what IC to hook up to drive the LED - I am still searching through micropower converters or sources, oscillators... I cannot really tackle anything smaller than a SOT-23..

My not-so-sound idea is to make something that do not require bulbs or batteries, and work for some 30 seconds at a time. I only need the SPICE solar model get an idea about circuit start-up and such before building something. I also tried to predict what max voltage may possibly appear on the output of the cap bank, and now I know - 6.2V. And yes, this is just a fun project for me. Thanks again!

[RJFreeman]

the Solar cell acts as a constant current source, so as when the capacitor is discharged it will draw all the current it can get and the terminal voltage will be 0 (at first) but as the capacitor charges, it will reach the point where it draws less current and the terminal voltage will increase until it reaches the open circuit voltage of the solar panel (whatever that is).

If I have a solar panel with a 6V and 16mA open values..

ok Open circuit current should be 0 (not 16mA) so I am not sure if 6V is your open circuit voltage, if I choose a 6V panel at near random from the Jaycar website:
http://www.jaycar.com.au/Ecotech/Solar-Power/Amorphous-Panels/1-Watt-6-Volt-Solar-Panels---Amorphous/p/ZM9020

this gives an open circuit voltage of 11V, so if you were using this panel your capacitor would charge up to 11V

This is kinda important because the caps are input to an IC with 5.5V absolute max

And fry your circuit....

If your open circuit voltage is 6V, then VF (Forward voltage) of your diode will decrease as the current through it decreases (as the capacitor reaches full charge) so the terminal voltage across the capacitor will reach the open circuit output voltage, of your solar panel

[abit]

Thank You!

I posted because I selected the specific PV panel for this, and I am not sure if it will do.

I have an LTSPICE model for this, and it shows that the PV cuts out at 16mA and drops voltage as it should..
The caps in parallel only serve start-up surge to the IC's inductor, so as you said, the voltage drops,
but I am worried that if the caps charge for too long they reach beyond 5.5V and the circuit fries.

Moreover, the caps are in parallel 10 or more plus soldered (I have fixtures to solder as many as I need) so cap ESR is basically non-existent...

Here is the LTSPICE model... If you or others could comment on this I thank you! Thanks for your reply!

Version 4
SHEET 1 5292 2260
WIRE 288 16 -48 16
WIRE 608 16 288 16
WIRE -48 48 -48 16
WIRE 288 48 288 16
WIRE 608 48 608 16
WIRE 288 144 288 112
WIRE 608 240 608 128
WIRE 288 288 288 224
WIRE 544 288 288 288
WIRE 288 320 288 288
WIRE 288 448 288 400
WIRE 608 496 608 336
WIRE 608 496 496 496
WIRE 288 544 288 528
WIRE 336 544 288 544
WIRE 432 544 416 544
WIRE 288 576 288 544
WIRE 608 608 608 496
WIRE 496 656 496 592
WIRE 544 656 496 656
WIRE -48 704 -48 128
WIRE 288 704 288 656
WIRE 288 704 -48 704
WIRE 416 704 288 704
WIRE 608 768 608 704
WIRE 688 768 608 768
WIRE 752 768 688 768
WIRE 944 768 880 768
WIRE 1104 768 944 768
WIRE 1344 768 1184 768
WIRE 1392 768 1344 768
WIRE 1440 768 1392 768
WIRE 1488 768 1440 768
WIRE 1536 768 1488 768
WIRE 1584 768 1536 768
WIRE 1808 816 1728 816
WIRE 1920 816 1872 816
WIRE 2016 816 1920 816
WIRE 2112 816 2016 816
WIRE 2192 816 2112 816
WIRE 1344 832 1344 768
WIRE 1392 832 1392 768
WIRE 1440 832 1440 768
WIRE 1488 832 1488 768
WIRE 1536 832 1536 768
WIRE 1584 832 1584 768
WIRE 416 896 416 704
WIRE 1808 928 1760 928
WIRE 1920 928 1920 816
WIRE 1920 928 1872 928
WIRE 1584 960 1584 912
WIRE 1728 960 1728 816
WIRE 1728 960 1584 960
WIRE 1536 1008 1536 912
WIRE 1760 1008 1760 928
WIRE 1760 1008 1536 1008
WIRE 1488 1040 1488 912
WIRE 1792 1040 1488 1040
WIRE 1920 1040 1920 928
WIRE 1920 1040 1872 1040
WIRE 1344 1088 1344 912
WIRE 1344 1088 1104 1088
WIRE 1440 1104 1440 912
WIRE 1920 1104 1920 1040
WIRE 1920 1104 1440 1104
WIRE 1392 1168 1392 912
WIRE 2016 1168 2016 816
WIRE 2016 1168 1392 1168
WIRE 1152 1232 1056 1232
WIRE 1248 1232 1152 1232
WIRE 1344 1232 1248 1232
WIRE 1440 1232 1344 1232
WIRE 1536 1232 1440 1232
WIRE 1632 1232 1536 1232
WIRE 1728 1232 1632 1232
WIRE 1824 1232 1728 1232
WIRE 1920 1232 1920 1104
WIRE 1920 1232 1824 1232
WIRE 1056 1280 1056 1232
WIRE 1152 1280 1152 1232
WIRE 1248 1280 1248 1232
WIRE 1344 1280 1344 1232
WIRE 1440 1280 1440 1232
WIRE 1536 1280 1536 1232
WIRE 1632 1280 1632 1232
WIRE 1728 1280 1728 1232
WIRE 1824 1280 1824 1232
WIRE 1920 1280 1920 1232
WIRE 1056 1376 1056 1344
WIRE 1056 1376 1024 1376
WIRE 1152 1376 1152 1344
WIRE 1152 1376 1056 1376
WIRE 1248 1376 1248 1344
WIRE 1248 1376 1152 1376
WIRE 1344 1376 1344 1344
WIRE 1344 1376 1248 1376
WIRE 1440 1376 1440 1344
WIRE 1440 1376 1344 1376
WIRE 1536 1376 1536 1344
WIRE 1536 1376 1440 1376
WIRE 1632 1376 1632 1344
WIRE 1632 1376 1536 1376
WIRE 1728 1376 1728 1344
WIRE 1728 1376 1632 1376
WIRE 1824 1376 1824 1344
WIRE 1824 1376 1728 1376
WIRE 1920 1376 1920 1344
WIRE 1920 1376 1824 1376
WIRE 1024 1440 1024 1376
FLAG 1024 1440 0
FLAG 2112 816 SCA_OUT
FLAG 688 768 PV_OUT
FLAG 944 768 PV_OUT
FLAG 416 896 0
SYMBOL voltage -48 32 R0
WINDOW 123 0 0 Left 2
WINDOW 39 24 124 Left 2
SYMATTR SpiceLine Rser=0.001
SYMATTR InstName V1
SYMATTR Value 8.35
SYMBOL diode 272 48 R0
SYMATTR InstName Solar_D1
SYMATTR Value 1N4148
SYMBOL res 272 128 R0
SYMATTR InstName SC1_R1
SYMATTR Value 480
SYMATTR SpiceLine tol=1 pwr=100
SYMBOL res 272 304 R0
SYMATTR InstName SC1_R2
SYMATTR Value 520
SYMATTR SpiceLine tol=1 pwr=1
SYMBOL res 272 432 R0
WINDOW 0 -80 37 Left 2
WINDOW 3 -49 70 Left 2
SYMATTR InstName VO_R1
SYMATTR Value 910
SYMATTR SpiceLine tol=1 pwr=1
SYMBOL res 272 560 R0
WINDOW 0 -80 37 Left 2
WINDOW 3 -54 67 Left 2
SYMATTR InstName VO_R2
SYMATTR Value 9.09k
SYMATTR SpiceLine tol=1 pwr=1
SYMBOL res 592 32 R0
SYMATTR InstName Solar_R1
SYMATTR Value 10
SYMATTR SpiceLine tol=1 pwr=100
SYMBOL pnp 544 336 M180
SYMATTR InstName SC_Q1
SYMATTR Value 2N3906
SYMBOL npn 432 496 R0
SYMATTR InstName VO_Q1
SYMATTR Value 2N3904
SYMBOL npn 544 608 R0
SYMATTR InstName VO_Q2
SYMATTR Value 2N3904
SYMBOL res 432 528 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName VO_R3
SYMATTR Value 100k
SYMATTR SpiceLine tol=1 pwr=1
SYMBOL diode 1808 832 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName SD1
SYMATTR Value 1N4148
SYMBOL cap 1904 1280 R0
SYMATTR InstName SCU1
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1808 1280 R0
SYMATTR InstName SCU2
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL schottky 1808 944 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName SD2
SYMATTR Value 1N5817
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res 1776 1056 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 0 56 VBottom 2
SYMATTR InstName SR1
SYMATTR Value 1k
SYMATTR SpiceLine tol=1 pwr=1
SYMBOL cap 1712 1280 R0
SYMATTR InstName SCU3
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1616 1280 R0
SYMATTR InstName SCU4
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1520 1280 R0
SYMATTR InstName SCU5
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1424 1280 R0
SYMATTR InstName SCU6
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1328 1280 R0
SYMATTR InstName SCU7
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1232 1280 R0
SYMATTR InstName SCU8
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1136 1280 R0
SYMATTR InstName SCU9
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL cap 1040 1280 R0
SYMATTR InstName SCU10
SYMATTR Value 10µ
SYMATTR SpiceLine V=10 Rser=0.02
SYMBOL res 1200 752 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName Limit_PV_R1
SYMATTR Value 0.024
SYMATTR SpiceLine tol=1 pwr=1
SYMBOL res 1568 816 R0
SYMATTR InstName Solar_J1
SYMATTR Value 0.024
SYMATTR SpiceLine tol=1 pwr=1
TEXT -32 192 Left 2 ;BOM Item #1:\nwww.futurlec.com\nPoly Blue cell -- SZGD5020\nPrice: $1.20 x 100\nEfficiency: 16.5%\nVmp: 5.38V\nVoc: 6.20V\nImp: 16.0mA\nIsc: 17.8mA\nDimensions (mm): 50x20x3
TEXT -48 -8 Left 2 ;NOTE 1: LTSPICE simulation for solar cells, NOT part of a BOM!
TEXT -40 -216 Left 2 ;CREDITS:\nThank you all!\n-\nwww.solarbotics.net\nWilf Rigter, original publisher\nEric Seale, page author\n.. and others.
TEXT 1112 1112 Left 2 ;Solar out (Pin 2)
TEXT -8 88 Left 2 !.tran 10000m startup
TEXT 1112 1064 Left 2 ;Pin 1
TEXT 1928 1152 Left 2 ;Pin 2
TEXT 1928 1088 Left 2 ;Pin 3
TEXT 1928 1024 Left 2 ;Pin 4
TEXT 1928 912 Left 2 ;Pin 5
TEXT 1928 800 Left 2 ;Pin 6
TEXT 1224 872 Left 2 ;JP_PV
TEXT 880 792 Left 2 ;PV Switch Panel

[RJFreeman]

Sorry I don't spice....
 

since your PV has a fairly low output current, the simplest solution would be to use a Zener diode to limit the voltage from the PV (as per dodgy paintbrush diagram below) - This will limit the maximum voltage across the capacitor, although again, as current increases the terminal voltage will increase so a 'nominal' 5V Zener may reach, say 5.5 Volts* so you may have to try a couple of Zeners to determine which one to use.

Given your numbers from before, if your maximum current is 16mA and you used, say a 4.7V Zener, then the power rating of the Zener Diode needs to be .016 * 4.7 = 75.2 mW (then I would usually say add a healthy fudge factor, but since the smallest Zeners I can get are 400mW this should be safe enough)

*I may be being overly pessimistic here, but it would still be good to test the characteristics of your Zener diode to be safe.

[abit]

I am hoping for a self-charging solar module with NO circuits to moderate itself, so this is a no-go - for one Zeners might not be for micro-power, as I read. Then, at this point, all I have is SPICE and a bunch of ideas and I am just here looking for more of the same. Thanks again!

[RJFreeman]

I am hoping for a self-charging solar module with NO circuits to moderate itself,

By this do you mean with no regulator? then you are going to need something that can handle a fairly wide voltage range.

for one Zeners might not be for micro-power, as I read.

well they can start to draw a few mA as they approach the knee voltage, which would potentially rob some power from the PV but it is a simple solution, (and this is where application requirements and engineering comes into play) sometimes it is cheaper/easier to use a slightly bigger PV to allow for this
- Also putting the Zener directly across the Solar panel and using a low leakage rectifier diode would stop Zener leakage from discharging the capacitor.

The alternative would be a switching regulator, but then you may need something that can boost and buck, something like a CUK converter, although boosting below a certain point is not really going to work, since there wont be enough current anyway, so maybe a buck converter with under-voltage shut-off, But then you seem to have also indicated you don't want a regulator?

So another possibility, probably the simplest and tried and proven technique, would be a solar panel, Battery and charge controller of some sort (to stop the battery overcharging)
It would probably be cheaper to buy a solar light and use the PV and battery from that than build something from scratch.

Of course without knowing exactly what your application is, it is difficult to know what to suggest 

[abit]

Yes, I mean no regulator..

The bottom line is:

The particular PV is a Voc 6.2V, Isc 17.8mA, Vmp 5.38, Imp 16.0mA as rated.
  - If the PV is rated at open circuit 6.2V such as this one, and if the PV "breaks": it is not going to break by swinging up to say 7V, charging the caps to 5.5V or higher, right? PVs should ALWAYS break by opening at least part of their circuits and decreasing their rated output, right?
  - So, there should be NO NEED, at least in my case, for a pre-regulator such as Zener or oh boy, another IC?

SPICE is giving me some 5.3-5.4V maximum at uAmps across a 1N4148, which is good enough for me, even if I scale this down to 10uF and give it minutes to charge.

  - Yes, the IC's inductor is going to drop a lot of the caps voltage, but not before the IC is at least partially switches ON, and the input is already saturated. Am I taking chances here to introduce the IC to higher than the absolute maximum 5.5V?
  - Would an xxxK resistor suffice to keep the PV in 'regulation' below 5.5V as it's possible maximum?

A bit more explanation, I have no idea what I am doing, I just need to see and understand the possibilities and the potential failures, and find some simple but efficient solutions. I might wrestle with the final equations later, but for now I see no need for that.

And - this is it, a micro-micro-power solar voltage reference able to put some 7.2V@500uA-1mA for as long as it can, absent of natural light. Efficiency at this point is not even an issue, but it should work and it should not ruin.

I really appreciate your responses, thank you!

[Seekonk]

Even having worked with solar many years, I am clearly not smart enough to follow this discussion.  The solar panel does not have to break for the voltage to rise, it just has to get colder.  Voc is specified at a specific temp and it will go up as it gets colder just like a diodes forward voltage.  I think a generation has lost something because of spice.

[abit]

Oh, Thank You!

This one is another fact I did not see! As my circuit gets colder, maximum potential voltage across the PV cell or the panel gets higher! I knew this but never so far considered it. Thanks!

Seekonk, you probably are an electrical engineer, and I am not. SPICE for me is what a telescope is for astronauts. Just to show for the rest of us what is like being so much closer to the real thing. A curiosity for me, a real thing for the astronauts or for the EEs.

Now, is there a chance when 0.45 Voc paneled cells get so messed up that voltage may rise beyond any expectations, independent of temp? Yes, No? Or PV cells or panels may never 'short-circuit' - even if there is a solar storm, EMP charge, whatever? They rather go ruined than exceed any voltage output limitations?

Please, this is important to me.

[suicidaleggroll]

I don't understand why you don't want a charge controller?  Any space saved by not having one will be wasted 10x over by having to oversize your panel by a ridiculous margin to keep the entire system from shutting down all the time.

In fact it's entirely possible your system will never turn on, depending on it's minimum input voltage threshold and efficiency.  You plan to draw ~7mW, yes?  Assuming an efficiency of 50%, that means your panel needs to supply 14mW.  If your circuit turns on at, say, 2V, it would need to draw ~7mA from the panel.  The short circuit current under full sun is only 18mA, so with even a little bit of shade you're going to be stuck in an infinite loop where the panel charges the cap to 2V, the circuit tries to turn on, it draws more current than the panel can provide, which drops the voltage below 2V and everything shuts back off, just to start over again a few ms later.  With a charge controller, even if the panel could only provide 5mA, it could do it at ~5V, which is 25mW - plenty to run the circuit.

[abit]

sucidaleggroll -

WHY would I want a charge-controller between a uA PV, a diode, and some 1000uF capacitors charging from the uA PV? I do not understand my own frustration around this idea, let alone yours.

When I use a PV the size of which is everywhere in pocket-calculators, I would like to count to 1000 elephants, and viola - the 1000u ceramic cap bank is charged to a maximum of 5.4V to start up an external boost DC/DC IC with 4.7uH... that is all. It's a boost-start circuit to a hungry inductor, nothing else, once the IC is humming it can feed off from 1.5V cells to a magnitude of other energy-rich sources, including larger PVs.

This is no super-capacitor or a large-scale deployement, but the 1000uF low-leakage ceramics should manage to climb to 5.4V no matter the conditions - as long as the 1000uF bank holds the last charge, the micro-power PV is going to keep stepping voltage up.

How else could I describe this? A milli-second generator that is smaller than a cigarette lighter, never needs batteries or sunlight, and sure to start up any IC?

[abit]

So, speaking strictly,

- in a circuit of PV-diode-cap will the caps ever charge to PV-Diode voltage, ending the charge cycle near, at, or above the voltage level that is balancing the charge (by the diode's reverse leakage, the capacitor's internal leakage, temp coefficient, and OTHER factors).. OR THIS IS NOT GOING TO HAPPEN?

The PV is the one here, for me to match, since everything else is certain. The target voltage is up-to 5.4V, as ICs with max of 5.5V for input tolerate by design voltages of 5.5V for start-up, for a very short period of time. By my design they do not tolerate 5.501V for any length of time.

The inductor for most boost ICs I read up, on the other hand, do not tolerate rapid voltage drops during startup, so the 5.4V from the start-up source is the best approach I came up with so far.

All I am really looking for is 'this will never work', or 'you need to think about this and that'

Thanks for reading. I am hoping to be a good sport here.

[RJFreeman]

in a circuit of PV-diode-cap will the caps ever charge to PV-Diode voltage

Yes the Capacitor will charge up to the PV terminal voltage minus the VF on the diode.
BUT:
A PV acts much like a constant current source (in this instance) so as the current drawn from it drops, the output voltage increases until it reaches the Maximum OC voltage (which you have given as 6.2V).

VF or the Forward Voltage drop across the diode, decreases as the current through the Diode decreases, and also as or if the Diode gets warmer theoretically approaching 0 as the capacitor reaches full charge.
The Vishay data sheet shows that VF is approaching 0.3V at 0.1mA at 90C  in fact it may be worth looking at the graphs (Fig 1 and 2) on page 2
( http://www.vishay.com/docs/81857/1n4148.pdf )

As Seekonk has reminded me, the PV will also increase output as or if it gets colder although this may be balanced by the increase in VF on the diode (but again that will drop as current draw decreases)

So In conclusion; your capacitor will charge up to somewhere near 6.2V and possibly even higher if you are in a colder area and while it may be that there is enough leakage in the capacitors or current draw from your circuit this is by no means assured and you would need to build it and test it under a range of possible conditions to see.

[suicidaleggroll]

So, speaking strictly,

- in a circuit of PV-diode-cap will the caps ever charge to PV-Diode voltage, ending the charge cycle near, at, or above the voltage level that is balancing the charge (by the diode's reverse leakage, the capacitor's internal leakage, temp coefficient, and OTHER factors).. OR THIS IS NOT GOING TO HAPPEN?

All I am really looking for is 'this will never work', or 'you need to think about this and that'

Thanks for reading. I am hoping to be a good sport here.

I think you're missing two things:

1) The load will not wait until your system is good and ready before it turns on (unless you add a voltage reference, comparator and transistor).  It's going to turn on the instant the cap voltage is high enough, which may be WAY before your panel is putting out anywhere near its rated power.

2) Panel power is generally rated at 1000 W/m^2 solar irradiance.  You are only going to hit this in a desert at high noon in the middle of summer with no clouds in the sky.  A typical sunny day is probably going to be more like 5-700 W/m^2 (cut your panel rating in half), and that's still at high noon outdoors in the sunshine.  Add a cloud or shadow in the mix, or use it in the morning/evening and you can cut your rated panel power by a factor of 10 minimum.  Bring it inside under artificial light and you can cut your rated power by 1000, easy, and that's WITH a good charge controller.  Your panel is only rated at <10x your load, which means even WITH a charge controller you'll be able to use it outdoors in the sun, and that's it.  Without a charge controller, my bet is you're going struggle to find enough light to get your system to turn on, even outdoors in the sun.  The lower the minimum voltage spec of your circuit, the worse it's going to be.

[RJFreeman]

I think you're missing two things:

Two or more, which is why abit may need to go from Spice land, pony up a few pennies (and the parts abit is talking about are not expensive) and take them into the real world and try a few measurements, and with what has been covered here, abit should have an idea of what to look for and what to measure....
 

[Seekonk]

I would use a TL431 in place of the zener in that simple zener circuit.  Low voltage zeners are a bit sloggy and you won't get just the right voltage with a zener.  Three leads and two resistors shouldn't be an issue.

For the edification and amusement of readers download LTC35882 data sheet.  An ultra low power converter that will work with piezo sound generartors.

[abit]

Yes, for a 5.5V IC I am going to need to put something between the caps and the input, and thanks for the TL431 tip! Especially because it is through-hole...