💬 The Harvester: ultimate power supply for the Raybeacon DK

[Mishka]

@Mishka said in 💬 The Harvester: ultimate power supply for the Raybeacon DK:

This all makes me think that there are basically two combos to choose from:

Not quite. After closer look I've found the AM-5610 outdoor panel of suitable size - only 25x20mm. The panel is capable to produce up to 18mW.

Other interesting panels are: AM-1606 as used on the Cypress BLE, 15x15mm, AM-1456 which is close to SolarBit by size, 25x10mm, and AM-1312 which is exactly of the same size the SCNE I have.

[NeverDie]

@Mishka I suppose the SC-3722-9 may be too big for your project, but it's worth mentioning because it performs pretty decently under indoor lighting, and you can extract them for cheap from $1 solar keychains, which are widely available. That's all subjective though. I'm not sure how they compare by the numbers.

https://www.aliexpress.com/item/32774648368.html?aff_request_id=d8e42a07ed7f4345b1133c9f4f249fed-1581813033932-00280-_rIgCIO&aff_platform=api&cpt=1581813033932&sk=_rIgCIO&aff_trace_key=d8e42a07ed7f4345b1133c9f4f249fed-1581813033932-00280-_rIgCIO&terminal_id=abc87087a3a64790a6abd96b2333811b

[NeverDie]

This diode is a bit expensive and too large for your project, but for experimental purposes it seems like the cat's meow as a blocking diode when collecting currents at tiny voltages: http://www.ti.com/lit/ds/symlink/sm74611.pdf

Just 26mv forward voltage drop at an 8a current, and just 0.3ua reverse leakage current at a voltage of 28v. Obviously, those numbers would be far less for the currents and voltages that we're dealing with. It seems to be very nearly an ideal diode, or at least the closest I've ever seen to that.

[Mishka]

Summarizing the work on the SPV1050 (irrelevant to other ICs mentioned in this topic) please let me publish revision 0.9 of the Harvester board. It addresses some issues found on the previous boards, and introduces number of important changes.

The most noticeable one is that the board now supports both boost and buck-boost DC-DC configurations of the SPV1050. After reviewing some tiny PV panels it was indetified that the maximum working voltage for a tiny panel is about 3V which means the boost mode is more suitable to do the job. Also, tiny high voltage panels (like some SolarBIT models) have very limited current capabilities and in low light conditions simply can't supply enough power in order to make the harvester chip work. It's also important to note that the cold-boot voltage for boost DC-DC is 0.55V which is only about 20% of maximum 3V voltage a panel can gain - please compare that with 2.6V and 4.4V respectively for some most advanced cells. Finally, if in the boost mode the SPV1050 supports TEG modules.

Therefore the BOM was adjusted to the boost configuration with the following thresholds:

Symbol	Parameter	Value
V_uvp	Battery under voltage protection threshold	2.4V
V_eoc	Battery end of charge voltage	3.1V
V_oc	Source open circuit maximum voltage	4.7V
V_mp	Maximum power point voltage	78% * V_oc

Please note, in the boost mode the SPV1050 will effectively set V_eoc = V_in for all V_in values greater than 3.1V which may cause damage to the battery or the nRF52 SoC. To prevent the negative impact please carefully consider the source.

If the only source you have is a high voltage solar panel, it's possible to adjust the R1-R3 resistors ladder (please refer to the SPV1050 datasheet) and switch the Harvester board to buck-boost mode as follows:

Hint: If the solar panel is really big (like 2W / 12V or so) and you don't need MPPT, it's possible to close the USB Charge jumper and connect the panel to VBUS and GND solder pads in order to employ the 3.2V USB LDO.

The MPPT fixed voltage ratio is set to 78% with resistors R2=2.2M and R3=8.06M. For a TEG module with MPPT ratio at about 50% just replace both R2 and R3 with 5.2M resistors.

Other notable changes included into this release are:

• Fixed some silk layer errors
• The MIC5205-3.2 LDO got missing input filtering capacitor
• The current limiting resistor between the SPV1050 and the tantalum capacitor has been removed
• Connection to the ground plane in some isolated areas was improved

And last, but not least, I'd like to thank the MySensors forum and in particular @NeverDie for his tremendous contributions. It's pure fun to discuss tiny boards with tiny harvesters working from tiny power sources :call_me_hand:

[NeverDie]

@Mishka Looks like a winner. :+1: Once you put it together I'll be interested to hear what the lowest light levels are that you're able to run it at and which solar panels/cells you end up liking the best.

I think there's a good chance it will outperform the eval kits from enOcean, Cypress Semiconductor, Cymbet, and others that rely on a high cold start voltage. In order for them to win they would need to harvest at a lower power than what your chip can manage but somehow also at the higher voltages and with enough power that their chips require, and I'm not sure whether or not those two conditions can be generated simultaneously by real world solar panels.

[Mishka]

@NeverDie Well, the SPV1050 has a nice set of features I need and offers impressive flexibility in a small package. However, when speaking about efficiency the AEM10941 seem outperforms it in every single bit.

The current design reached some level of stability so I think the AEM10941 is what I should try next.

[iiibelst]

@Mishka said in 💬 The Harvester: ultimate power supply for the Raybeacon DK:

The current limiting resistor between the SPV1050 and the tantalum capacitor has been removed

Why removing the resistor? Why not placing it after the (optional) Tantal to protect a downstreamed Batt or Cap?

[Mishka]

@iiibelst There is one. The R7=549Ω is limiting current between the tantalum capacitor and the extension socket. Its purpose is to keep it under 2mA for ML2032. You can bypass it with relevant solder jumper on the board bottom.

The dropped 50Ω resistor was previously located between SPV1050 and the tantalum capacitor.

[NeverDie]

@Mishka said in 💬 The Harvester: ultimate power supply for the Raybeacon DK:

The current design reached some level of stability so I think the AEM10941 is what I should try next.

Yes, at a 3 microwatt minimum, that chip may be very tough to beat. It has the same 380mv cold start voltage as the ADP5091, but it requires only half the energy. With these tiny solar panels that extra margin might really make a difference under dim indoor lighting conditions.

I guess it's no accident that the AEM10941 is the newest chip. Perhaps it's the constant improvements in cmos technology that it leverages. In which case.... we can probably look forward to even better chips in the future! For sure solar cells and panels continue to improve their efficiency. The markets are finally big enough to support the required R&D for continual improvement. And the mcu's and radios are constantly improving their efficiency so less power is required. It's great to be in the nexus riding a few waves like this, where we can get the benefit of multiplying the improvements together.

[Mishka]

@NeverDie That's all true.

Please also note the AEM10941 can regulate up to 5V of input when ADP5091 upper limit is 3.3V.

Again, when speaking about BLE, a beacon (as a low-power application example) has to advertise at least once every 1000ms to be generally usable. For nRF52840 it approximates to about 50µW of power consumption. By adding up some microwatts of the harvester itself it may be safe to expect a panel should be able to produce 60-70µW of energy in average. In turn, this means that those 3µW or 15µW are rather an edge scenario, and there is must be a timeframe when the system can collect all the required electricity. Such, when running from a daylight it should be expected that in February the harvesting will be efficient for at most 8 hours a day. The system must be able to offer minimum (24h/8h)*70µW = 210µW during the light period of time. For the reference, a couple of my IXYS panels can provide only about 150µW when located in 1m from window on the north side.

From the experiments, to me it currently looks more not about possible minimums, but rather about higher efficiency at nominal values. But I admit the difference between 3µW and 50µW doesn't look big either. The sleeping current of the nRF52 is already less than 3µW, so perhaps some time the source and the load can converge.

[NeverDie]

@Mishka said in 💬 The Harvester: ultimate power supply for the Raybeacon DK:

Again, when speaking about BLE, a beacon (as a low-power application example) has to advertise at least once every 1000ms to be generally usable.

I read in a couple different places that the maximum bluetooth advertising interval is 10.24 seconds (i.e. 10x your assumption) so if the rest of your math is right that should provide ample headroom for being bluetooth compliant.

[Mishka]

@NeverDie Yeah, there are also scenarios when a beacon advertises only when it was charged enough - it may wait for several hours before send a packet. But that's rather uncommon application. A typical beacon usually advertises every 100 ms to 1000 ms - this way it can be located quick enough.

[NeverDie]

Closing the loop: today I finally received the sunpower solar cell, so I was able to take a closer look at it. Basically, the traces on the back are interdigitated. So, it looks as though it could be cut along the horizontal axis (if, say, the connection pads are on the left and right) almost as narrow as whatever you might want to. However, it would be ruined if you were to cut along the vertical axis: one pad would remain fine, but all the traces to the other pad would be severed. Maybe in theory they could be re-attached to a new pad with a lot of careful soldering, but that doesn't seem very practical, as the pitch between traces is quite narrow. On the other hand, if one were to use a custom flex film pcb with connection traces that aligned to the severed traces, it might be possible, but still a nontrivial amount of work.

Looked at from the point of view where large surface area is OK: one of the nice things about these cells is that they are reasonably inexpensive considering their 5"x5" width and height, and yet they are quite thin and still easy to connect. However, I suppose they maybe should be coated with something to protect them. A 2K automotive epoxy spray would probably be ideal, but perhaps even a hard automotive acrylic lacquer would be sufficient, as either should be both non-yellowing and moisture proof. Unfortunately, not much seems to be written about what types of coatings work best. Obviously, the commonly used chinese epoxy solar cell coating would be a poor choice, as that stuff degrades under uv and yellows/browns and clouds up quite rapidly.

[Mishka]

@NeverDie Interesting. I think the cell may be carefully cut with laser and then properly remetallized. Perhaps can be done with a typical tin based solder paste with some proper flux (I don't know, originally some kind of silver paste is used). Fixing the cell itself into epoxy should be easy.

The nice thing about the process is that it should be virtually possible to create cells of arbitrary shape. However, in order to get usable voltage it might require to build a panel.

[NeverDie]

@Mishka On second thought, if you were cutting it to a small size then there wouldn't be many traces remaining to be reconnected, so from that point of view it might actually be practical.

For me it's academic because I don't own a laser cutter, and I have no idea what kind of power would be required to cleanly cut one of these cells even if I were to buy one for that purpose. I'd be interested to know though. Even 20 watt lasers are pretty cheap these days. Hooking a laser up to a CNC, which I do have, to execute the cut would be fairly easy.

[Mishka]

@NeverDie My nearest laser service costs about 3-4 dollars for one running meter, can cut 2 mm steel, so never thought about that. On the other hand, with enough number of passes it virtually should be possible even on a DIY CD-ROM laser engraver, especially if mounted on a CNC which is usually more precise than lasers.

I'm expecting that at least three crystals will be required to gain 1.5-2V. For a circle, it sounds reasonable to cut three or four sectors of equal shape. Maybe a 3D printed pallet can be used with top layer protected by epoxy. But I think for the best result additional metallization will be required anyway.

[NeverDie]

@Mishka Since you could convert one large cell into lots of tiny cells, it might actually be cost effective.

I once looked into directly etching the copper on pcb's using a laser mounted cnc, but apparently that requires a much higher power and more expensive laser than what's commonly used by hobbyists. AFAIK, simply running a low power laser over the same isolation traces over and over with just a low power laser won't get you anywhere. I'm guessing that's because of both the copper's reflectivity as well as pretty excellent heat dissipation to the surrounding copper. It may be that a solar cell wouldn't be as difficult, but I couldn't say. For sure your cutting service's laser should be able to handle it though. Please do post how it goes if you decide to pursue it.

For POC you could cut the cell using just a box cutter or something like that. What happens is that the cell shatters near the cut mark, but enough is left over that the cell still works. So, it's not really the proper way to do it, but it could be done, at least for larger cells. It's hard to know a priori how far the shattering/cracking might travel in a small cell. Maybe not much useable area would be left. Or maybe there would be. I guess that would require experimentation to find out. I only know what I saw in this youtube video:
Cutting Supower Maxeon Solar Cells? - Mikes Inventions – 06:37
— Mikes Inventions

His was just a rough and ready test to see what would happen. Perhaps cutting it on a circular saw with a suitable tile cutting blade, tightly sandwiched between lots of rigid support might cause less shattering/cracking. Or perhaps borrow techniques used for cutting thin glass. Perhaps multiple passes with a diamond drag bit on a CNC could do it with minimal shattering/cracking: https://www.amazon.com/Diamond-Spring-Loaded-Engraving-Degree/dp/B07F9L62C3/ref=sr_1_6?keywords=diamond+drag+bit&qid=1582661720&sr=8-6
I think that might stand a decent chance of working. However, aside from a POC, it's easy to see why a laser would avoid these problems altogether, and without producing dust.

[NeverDie]

Art Resin tested a large number of different epoxies, and it seems that all of them yellowed to some degree over time, but some a lot more than others:
Epoxy Resin Yellowing Third Party Testing from ATLAS Labs – 01:46
— ArtResin

Of course, since it was a test designed to make Art Resin look good, perhaps they omitted epoxies that really do never yellow. I just don't know which ones those would be. Eight weeks, which was the limit of their study, doesn't seem like a particularly long time.

[Mishka]

@NeverDie Wow, nice collection! How do you think, may it be reasonable to glue cut cells to a quartz or glass base? Would it compromise effectiveness?

IMHO the right way to cut them is either laser or high speed CNC. Also, CNC cut crystals may require extra polishing.

Just asked a couple of local vendors for a single cell, waiting for their reply. BTW those cells are usually of 18-19% energy efficiency, so the only way to beat Amorton or IXYS is to cover larger areas.

[NeverDie]

@Mishka said in 💬 The Harvester: ultimate power supply for the Raybeacon DK:

Also, CNC cut crystals may require extra polishing.

I think polishing would probably damage them. These cells are different than generic monocrystaline cells. Allegedly, at a microscopic level, they are built using tiny pyramids to increase their surface area. I can believe it, because when taken out of the package they look a bit like velvet. For that reason they apparently scratch extremely easily. The two that I received were in their raw form and totally unprotected, so I am right now in the middle of applying layers of an acrylic lacquer to them as a guard against scratching the active surfaces.

A water clear urethane coating might have been a better choice, as it's probably harder, but acrylic lacquer is all that I had on hand. I hope to handle differences in co-efficients of thermal expansion by coating both the front and the back equally. Otherwise, it will probably warp.

I soldered the dog-bones to them. I used rosin core solder, because that's all I have on hand, but next time I think I would use pure solder without the rosin, because I'm not sure whether the resin will interfere with a protective coating. I'll have a better idea about that when I finish coating this batch. Because of the cell's fragile nature and tendency toward scratching, I don't have the guts to clean off the resin with IPA without a protective layer in place. Perhaps I should, though, after the coating on the front finishes curing, and before coating the back of it.