This episode examines how energy harvesting reshapes the iot data platform for smart cities by removing batteries from the equation. Jérôme Vernet from Dracula Technologies explains how ultra-thin, printed organic photovoltaic films optimized for indoor light can power low-power sensors that feed modern iot data integration platform tools. We discuss iot data ingestion best practices when devices must operate for years without maintenance, and how LoRaWAN and other low-power protocols support scalable, battery-free deployments. The conversation also touches on what is privacy and trust in iot data platform design when billions of events come from always-on labels and tags. If you’re planning the best database to store iot data or comparing iot data analytics vs network analytics for large fleets, this episode offers practical context from the hardware layer up. Listen to understand how energy harvesting changes both device architecture and long-term IoT operating models.

Jerome on LinkedIn

Dracula Technologies

• Helium Global IoT Coverage - Want to know if Helium coverage exists where you need it? Check out this map!
• Helium Foundation - The Helium Foundation's IoT Working Group (IOTWG) has generously provided support for the first 6 months of shows, please go check them out and consider using the Helium LoRaWAN as a primary or backup on your next deployment. With over a quarter million gateways deployed worldwide, it's likely that you have and can use Helium coverage.
• Support The Show - If you'd like to support the MetSci Show financially, here's where you can donate on a one-time or an ongoing basis. Thank you!
• MetSci Show - If you'd like to use our IoT or AI Data Value calculators, or you'd like to contact me, the MetSci Show site is the best way to do it.
• MeteoScientific Console - Use LoRaWAN - The MeteoScientific Console allows you to use LoRaWAN today. As long as you have Helium coverage (and you probably do, about 90% of populated areas in the world have a gateway within 2 miles), you can onboard a sensor. You can always check coverage at https://explorer.helium.com and switch to the "IoT" tab in the top right.

Today's guest

2

on MeteoScientific's

The Business of LoRaWAN is Jerome Vernet,

3

VP strategy and co-founder of Dracula

Technologies, a French deep tech company

4

turning indoor light into usable power

for IoT devices.

5

Jerome brings a rare mix of industrial

manufacturing experience and ultra

6

low power design thinking,

and in this conversation,

7

we dig into what it actually takes to move

8

from battery powered sensors to energy

harvesting LoRaWAN deployments.

9

We talk about how Dracula evolved

from a small R&D lab

10

with a glove box into a company

with large scale, clean room production.

11

Why replacing batteries becomes

a financial decision at scale, and why

12

designing for battery free LoRaWAN means

rethinking devices from scratch.

13

Jerome also explains

where energy harvesting breaks down,

14

how ultra low power

electronics unlock new markets,

15

and what role autonomous sensors

will play in future edge AI systems.

16

This episode is sponsored

17

by the Helium Foundation and is dedicated

to spreading knowledge about LoRaWAN.

18

If you'd like to try Helium’s

publicly available.

19

Global LoRaWAN for free and support

this show, sign up at metsci.show/console.

20

Now let's dig in.

21

In conversation with Jerome Vernet.

22

Jerome,

thanks so much for coming on the show.

23

I'm excited to have you here today.

24

Thanks for the invitation

and glad if I can bring some nice insight

25

coming from a, you know, energy

harveesting devices, powered.

26

Let's say for any kind of filter

application and so on.

27

So. Yeah.

28

Glad to be with you.

29

Yeah, I first saw Dracula.

30

I first saw it just,

I think a year or so ago in Barcelona,

31

and you guys had the coolest card.

32

Most people will just listen to this.

They won't see the video.

33

But I'm holding up this card

that's got these little bat wings on it

34

that are done in this little solar panel.

35

And that's what Dracula does in

these kind of indoor solar panels.

36

And a little copper

37

connector is just like, it's

the coolest business card I've ever seen.

38

So that's

that's like a nice, nice way to start.

39

Let's see, how did you guys start?

40

You guys are based out of France.

41

Walk me through how long you've been

around, what you're doing, how it's grown.

42

Yes. We are all located in Valence,

it's the southeast part of France.

43

And early on,

44

It's a deep tech company,

so we started from scratch from an idea.

45

So Bris my partner, was crucial, the CEO.

46

He started the company

just getting that idea of printing

47

something different than traditional

graphic inks, because Valence

48

is very well-known and famous for graphic

printing and industries.

49

The big player worldwide

which is Image part of Domino group,

50

which is all printing

51

for on the eggs, on the bottles

and so on for for the food mostly.

52

And the idea was,

so can we use, this kind of equipment

53

to print something different

than traditional graphic inks?

54

And there it has been a European

project later and for two years,

55

at the end of the two years,

the major company decided not to proceed.

56

And Bris, my partner,

say there's something here, I do it.

57

And I started the company

58

Just then they started

first with many years,

59

you know, R&D

60

and I mostly in the universities

and the resources we have as a, a CNRS,

61

we chose center development

and will be able to bring some real

62

activity in Valence in 2016 here,

where we start getting and hiring people.

63

So at that time we were

64

when I joined the campaign in 2016,

we are four people, we are now 47,

65

and we started from a very small R&D lab

with, glove box and so on.

66

And now we have a clean room

of 450 square meter

67

and adding a new one of 200 square

meter now.

68

So that's a huge project.

69

Okay. Yes.

70

So from glove box to cleanroom,

that's that's pretty good.

71

That's a good move.

72

That has been a very challenging move.

73

But we made it short. Yeah.

74

And for folks who have kind of missed it

so far, what you guys do is you make film

75

that takes energy from light and turns it

into something that IoT devices can use.

76

That's kind of the simplest way

to explain it.

77

That's exactly this, basically.

78

And why drag lighter solar for indoor?

79

We didn't love the name Dracula

because we don't like the sun.

80

We are much more efficient in the indoor

light indoor environment.

81

So we use a graphic printer, traditional

graphic printers, and instead of printing

82

basic inks,

we develop our own formulas and the blend.

83

And we are printing

these functional inks to generate the,

84

the PVC, the OPV, organic photovoltaic

and all the products are carbon.

85

They no rarest, no heavy metals.

86

And how thin is the film?

87

I'm just getting into this

kind of thickness measuring phase.

88

So tell me anything.

This stuff is the complete device.

89

Thickness, about 300 micron.

90

300. You said. Yes, 300. Okay.

91

And only the biofilm

to encapsulate the product.

92

All about 200 micron.

93

Okay, so 500

total. The main thickness. Yes.

94

So no, no.

95

Among this, among the 300,

you have 200 already for the biofilm.

96

So all layers the different printed layer,

the four different layers in two ways.

97

Android micron remaining between ten

that we speak about nanometers.

98

That's why we need clean room

also to to print this.

99

Yeah. Yeah. Okay.

100

So these things are super thin.

101

They work best indoors.

I hadn't picked up on the.

102

The Dracula doesn't like the sunlight.

103

I'd only picked up

on the kind of vampire thing and

104

and drawing the energy out of the room,

so that's cool.

105

Er, yeah.

106

For the folks listening to this,

they're into LoRaWAN.

107

If you had to explain in one sentence

or maybe a couple more, why battery

108

free IoT matters to the LoRaWAN ecosystem,

what would you say?

109

I would say that everybody just,

going to big back to backward back.

110

Ten years ago, 15 years ago,

everything was on grid powered.

111

Nobody care about energy with

you was just on grid.

112

Later on, we say it would be nice

if I want to add one more.

113

It's always becoming

tricky to add a new sensor, a new IoT.

114

And so I need to put some cables,

blah blah blah. Not that easy.

115

So saying let's go wireless.

116

So we've been wireless in that time.

117

And yes, but we had another wire

which is the battery itself.

118

That means when you use a product

you have to replace a battery.

119

And the something very important,

it's when you display and you

120

spread around the building, a few sensors,

tens of sensor, you can deal with that.

121

So it's not a big deal.

122

Now if you think about it's

pretty much more product.

123

We speak about thousand, 10,000,

millions of units.

124

You can't imagine you the battery.

125

And I would take the example, the basic

one the 12 on the remote control.

126

I'm home.

127

I'm not battery

anymore on that remote control.

128

There's no real pain.

129

I can swap it.

130

I can get one

used from another, remote or change it.

131

And we figure out later.

132

Now, as all the IoT are using and mostly

133

using a lower protocol

and so on to transmit data,

134

because the keys are data,

the hardware is only a way to get this.

135

But without a decent

136

and efficient Arduino, reliable hardware,

there's no data available.

137

Yeah.

138

So that's why we've seen

because the electric consumption went down

139

so that now we are needing less

and less power and we're able to collect

140

more and more energy.

141

With all the energy

harvesting technologies around,

142

we have the light,

which is what we do with Dracula.

143

You have the RF, you have the Peltier.

144

With temperature, you go to vibration,

you have many other opposing technologies.

145

And combining them all

now to power autonomously

146

many application and the main part

of the consumption in an IoT

147

is not the sensing itself, it's

transmitting the data.

148

So that's why being able now

to collaborate and work on some

149

the project we developed also with a

we have a demo working with design

150

with Laura, it's also with Murata EP's

and different of the large players

151

in that market to work on independent

high sensors that would be able to

152

get the vision in the case, use the hand

and to transmit that information

153

in a in a real time, in very consuming,

very low power.

154

So in that case, it's a lower one,

let's say short shortened distance.

155

That makes sense.

156

And that's we label

you working within our job is to inform

157

this kind of application.

158

Got it.

159

So I mean it's it is a real shift

from where we were ten years ago

160

and saying, look, there's

there's so much energy floating around.

161

Let's harvest some of it. Yep.

162

And use it for these things.

163

And that eliminates

most of the battery requirements.

164

But there's still got to be some where

165

where it is kind of battery free IoT

break down in the real world.

166

What would you say?

167

Or some assumptions

that people just get wrong?

168

No. The funny thing is that now

using these kind of full power

169

electronics, combining this, combining

these kind of a new feature with energy

170

harvesting power solution

bring you to unlock new markets.

171

If we know what I speak one about,

let's say, for example, retail tracking,

172

logistics, basic one everybody

now is aware about to everything.

173

We all know how everything is working.

174

It's a passive tag.

175

And you send some some some waves RF

and you give the signal I'm here now

176

what we do with these kind of new labels

with new communication protocol.

177

Lora could also be the alien.

178

So on the I also emit able to transmit.

179

That means we extend the reading range

and you use some existing infrastructure.

180

So that's why now we are looking

new application, a new feature.

181

And again, because everybody's looking for

ways more and more and more data.

182

The data is key.

183

And again

nobody take care about the way itself.

184

So the fact that we able to team up

with large corporates such as Lora

185

so same tech and also player

big player would way to find new solution.

186

And the push the boundaries

of traditional limitation of batteries.

187

Because we've seen and still true.

188

If you take your basic remote control

and you check the thickness of

189

a remote control, the main space is given

by the battery itself.

190

Yeah, the Tripoli.

191

Now, if you remove it and you go in

very thin, you don't need all that space.

192

So you also able to reduce the overall

old format and consumption

193

of plastics and also product when you're

getting something much thinner.

194

And that the interest also of all

technology of OPV, which is also flexible,

195

is to be able

196

to follow the deformation and go

and from the shape of the product itself.

197

So this is an looking

mostly new application.

198

Do not consider this to charge your phone.

199

Many people ask me any time, oh great,

and you mentioned the call that you have.

200

Yeah.

201

Oh if I have somebody on my phone

I, we we made that.

202

We made the exercise. Just do the math.

203

If you want to recharge your phone

with something as big as the one you have

204

on your under the card, we shown you

if we get two years 24 seven okay.

205

So they're not super efficient for space.

206

So you've got to optimize for power.

207

But they work as long as there's lights on

208

and everyone

you have human, you have lights.

209

Yeah yeah yeah we do

we do like to be able to see that's.

210

Yeah, that's a nice thing about humans

or the important thing about humans.

211

Let's see.

212

You talked a little bit

about reducing plastic but it's it's

213

kind of blunt to see it.

214

But most see folks

don't really care about sustainability.

215

They care about dollars at what scale.

216

And you talked about going to thousands

first just like one remote control.

217

Now what scale does

removing batteries become

218

a financial decision

rather than a sustainability decision?

219

I would say that the first step

as being the fancy

220

and kind of nice to have,

and it was a system and it was green.

221

So we've come back

five years of six years ago.

222

Oh, that's a fancy stuff.

223

It's it was a nice to have reason

224

why we have ten many times the year for

let's to the consumer market.

225

I'm glad but I will not change the world

by getting one of the device in my,

226

in my room.

227

Now, if I'm using millions of sensors

228

embedded with millions of batteries,

then rechargeable disposable thing,

229

you have an issue

230

and that start becoming a real pain

because you have to take care of this.

231

Now the manufacturer need to figure out

how you will recycle the product

232

at the end of the life.

233

So that's the biggest issue.

234

Also something very important material.

235

We're using our carbon based product.

236

No rehearsed movie metals fully recyclable

is he made

237

this kind of a PVC is made at 99%

plus or 50.

238

So you can recycle the same way.

239

A traditional plastic bottle.

240

Now coming back to

how much we unlock, let's say for the CFO,

241

if you consider that the battery cost

ten cent for huge

242

volumes, the ten cent

and this my device cost $1.

243

Hey guys, your ten starting ten times

more expensive.

244

What do you want me to put?

245

I don't I agree it cost $0.10.

246

Now how much does it cost you

to change one of these battery

247

to send someone there to go?

248

I know maybe on the stair

to unlock the sensor.

249

Remove it, change

250

it, swap it, take it with care

about the ten cent of the battery.

251

Plus you have that

the first loss is array.

252

That means how long it will cost me.

253

If I have to swap to two batteries

in order for the product to learn.

254

Yeah.

255

So two way

to consider the retrofitting the battery.

256

You do it as anticipation.

257

Proactively say okay, every two years

I change the battery.

258

Maybe your friend your way.

259

Batteries are still full of energy,

so you're wasting energy or you want

260

you make it.

261

Let's say it's a corrosive maintenance for

my device is not blinking anymore.

262

I need to go there.

263

Maybe you lost much more money

by not having the data available.

264

So this is another saving.

265

Now we speak about battery

and embedded on a product with a new,

266

let's say, regulation for sending overseas

lithium batteries and so on.

267

You have to make specific paper works.

268

You have to follow specific rules.

269

This is costs also additional cost.

270

So that's why 1 to 1 battery this.

271

No. Now you speak about the

how many times you will use it.

272

One year two years for years

four years, five years old.

273

Technology is guaranteed 8 to 10 years

indoor 8 to 10 years.

274

And it is not binary.

275

It's not on or off.

276

No, it will decrease slowly

and that when you consider 8 to 10 years,

277

is because at that time we still get 80%

of the initial performance.

278

But we consider that the more efficient

as initially but still have 80%.

279

Okay, so after 810 years, 80% efficiency

okay.

280

So it's not going down that much.

It could run it.

281

You know depending on how you design it.

282

You we take a little bit more time.

283

Maybe you trust me to data.

284

It was every minute.

285

It could be every minute and five seconds

okay.

286

But you'll get something done. The road.

287

Okay.

288

So if I'm an engineer

and I'm designing a LoRaWAN sensor today

289

and I want it to be battery free,

what are some design compromises

290

I'd have to accept

or how should I think about it?

291

You have to restart from scratch.

292

It's the same thing

293

that if I go with my old fashioned

Peugeot car to a Tesla dealer.

294

Okay, give me one of your electric engine.

295

The guy will say thanks.

296

Nice thinking in my company,

but no chance.

297

It's only a matter of the engine

as a matter of the overall conception.

298

So you have to restart from scratch

299

because many people try to retrofit

quite easily.

300

Oh, this is a battery powered device.

301

Just plug your PV to replace it.

302

We do the job.

303

But the thing is, it's

a complete different way of thinking

304

when you speak about battery.

305

It's a tank full of energy, that tank

full of energy as may be untried inside.

306

And any cycle you will consume

.00 1% after many, many cycle

307

you reduce the volume of that,

the capacity of that to tank.

308

But when it's empty it's anti you.

309

My solution I will never get Android.

310

It's huge Android of energy.

311

I will give you 00.1.

312

Hey that that's not that much.

313

But that's good enough to the job.

314

I'm giving you this any time you need it.

315

So it it's that's way to a completely

different mindset to rubbing my.

316

And that's why we support

also with by sharing some redesign

317

just to help the people thinking

about the application.

318

Exactly

319

is not big is not always better.

320

And mostly for that reason

because if you take many IoT

321

and I have different in mind,

the form factor

322

is given by the battery

or the size of the battery itself.

323

I get a bigger battery

so it will be in mobile key.

324

So mobile key means more expensive stuff

to recycle, more amenities inside

325

and on the road.

326

The real efficiency is it

what I'm looking for.

327

Not that much.

Got it. Super cool to think about.

328

Yes design completely mindset.

329

And that's a funny thing that we have

a team in Dracula, but it's existing in

330

many of our partners

active in the energy harvesting market

331

that are expert in ultra ultra low powers.

332

I remember the first time

I was presenting that kind of DMCA.

333

I'm getting Android of my current people

look at me milliamp no micro.

334

What do you all need to do with Michael?

335

Now we add some customer

requesting about Nano

336

for that application.

337

We are using a BLT,

an example chip use worldwide

338

the consumption 650 nano and

339

nothing. Yeah nothing.

340

You don't need the bulk of your stuff. So.

341

And good enough to we to power

with 1% of that with a tiny little tiny

342

little piece of of this film

that is maybe for my for our American

343

listeners, about the size of a 50

cent piece, a half dollar.

344

Correct. Okay. Yeah.

345

Or I mean, you could cut a dollar

bill in half and get a little bit more.

346

Okay, let's wrap

this up with this idea of AI at the edge.

347

Do you think that battery free devices

will eventually do kind of meaningful

348

edge intelligence will be

will they be able to power that

349

or are they just constrained

to a simple sensor rule?

350

No. All AI is related to data.

351

And to get that out again, I get back

on what I told you a little bit earlier.

352

You need hardware because that's

where we collect the real data.

353

And after you'd be able to crush

all that data to to get a very agile AI.

354

So that's why sure, we only pouring

sensor, but sensors on the road are key.

355

You want to sense the temperature.

356

You want to send the location,

the humidity.

357

They want to sense

how many people are in the room you want,

358

see how many people when there

to get all your calculation, and so on.

359

So we are able to answer one

part of the answer, but not all for sure.

360

But we know where we are

good at where we are not good.

361

Right.

362

Don't don't even

363

you asked me to power your car

or even your phone and that does anything.

364

No, no, we are good at very ultra low

power weights.

365

Relevant.

366

And that's that's the most important.

367

And what we've seen on the on the way

with, with Laura for the past five years

368

has been to showcase

that many applications can be powered

369

autonomously, or we can even consider

to recharge the battery.

370

Many people are not considering

the leakage current.

371

The leakage current are huge.

372

When you do, you keep your TV on.

373

In sleep mode is not really sleep.

374

There's a vampire drug? Sure.

375

Let us manage that for you guys.

376

Yeah. Oh, that's that's super cool.

377

So there's this also this space

where you have your films

378

working alongside a battery

or maybe your super cap or whatever.

379

Exactly.

380

That's a combination.

381

But to get the perfect answer,

if everybody agrees on the way

382

we are wrong and that's exactly

also what the gain lower

383

I can demonstrate with a TTN show

last year in Amsterdam,

384

it was the first time ever

that they where they invited

385

all the different modes of protocol

communication protocol to team up,

386

and by doing so they unlock many, many,

many applications

387

and bringing new option, new solution

and opening new market.

388

That's that's huge.

389

And that's a great move for me.

390

Yeah.

391

Yeah yeah. It was it was good.

392

They called it the Things Network

and not the LoRaWAN network

393

from the beginning

because it opened the door to,

394

you know, ten years later, being able

to connect with all the different things.

395

I mean, you've seen

a lot of a lot of different devices.

396

We talking about some cool stuff before

the show with just looking at gestures.

397

And if you hold up one finger,

398

it says 1 or 2 fingers, two,

two and three fingers three, etc., etc..

399

What are some of the coolest

or most surprising

400

use cases

you've seen for folks using Dracula?

401

We had a request

from one tech manufacturer that has just

402

because we mentioned about looks, we speak

about luminosity, we speak about Lux.

403

And one sun for solar is Android 1000 lux.

404

So we are working

405

with some New Zealand

and traditionally OPV are powered.

406

And we make the calculation,

407

the measurement at 1000 lux,

which is 1% of the overall sun.

408

The guy said, you said that

you are able to go down to 50.

409

So yes, we go down to 50.

410

You say, okay, what about ten?

411

Yeah. What about 1 or 2 lux in it?

412

No, we can let's do the, the the the test.

413

We run the test. And that one.

414

Lux, we've seen that our technology

was able to generate the voltage.

415

Something crazy traditional PV

that normally generate 2.5V

416

was able,

in that case to deliver 1.1V at one lux.

417

Something crazy.

418

And the guys at great

at good enough for me

419

because my application require nano amps.

420

So nano arms and E5, the voltage,

which is the most important part for me.

421

I do the job.

422

That has been the craziest request we had.

423

It wasn't, it wasn't. How big

can you get? It's how small can you get?

424

And do you see this?

425

I mean, imagine

this gets used in mining companies

426

where there just is

not a lot of light underground.

427

No, no,

because even underground in a mining

428

or you have light everywhere,

even in the, in the mine

429

that are not in use as a safety feature,

they need to get light always on.

430

Sure.

431

So it's good enough to get enough power

to send data temperatures.

432

Okay. There's nothing

there's no special gas and so on.

433

But in that case UV was for complete.

434

It was for label.

435

That could be work in no lights,

roughly no light.

436

And it was a pretty dark warehouse

or truck or something.

437

Exactly.

438

And it was the same thing

as a kind of a combination of technologies

439

and say, I need a little bit of energy.

440

What I can get at one looks,

we say nothing.

441

First say, let's try it.

442

And we discovered

it was still active. Yeah.

443

Oh that's that's right.

444

Is there anywhere

445

where you guys want to be able to explore

that you haven't explored yet really

446

like, oh, this you are going to be doing

in the next whatever, one, two, five years

447

in the different verticals.

448

We were active the first one,

449

and it would come back to you

about the initial question

450

about the consumer market

and so on, which is for me, something

451

interesting to go green,

and it's something nice to consider.

452

But more than the green stuff,

453

I would say that

green it's additional value we need to do.

454

Make sure you are to be reliable

at the right price available,

455

and if we are green,

that's cherries on the cake.

456

The different market we want to address.

457

And there's something that we see.

458

It's more and more basic sensor

we can of install and forget device.

459

You leave it there and after

460

after maybe ten years or whatever

the product will it will disappear

461

because it's it's

basically it will be made on material.

462

We are thinking and doing some testing

on a new materials paper

463

or some of their normal patches

disintegrate.

464

Oh, that's amazing.

465

Disintegrate to make sure that okay,

at the end of the road, nothing special,

466

just the kind of application

that we are dealing with.

467

Oh, that's so cool.

468

Right on.

469

Well, Jerome,

thanks so much for carving out time.

470

I know you're super busy.

471

I appreciate you coming on

and telling us about Dracula.

472

My pleasure. Thank you.

473

That's it for this episode

of The Business of LoRaWAN.

474

If you want to go deeper

and actually deploy devices,

475

the MeteoScientific

Console is the fastest way to do that.

476

And honestly, it's

also the best way to support the show.

477

When you use the console, you're not just

listening, you're participating

478

in the same real world LoRaWAN work

we talk about here every week.

479

You can get started with a free trial

480

at MeteoScientific.com Huge

thanks to the sponsor of this show,

481

the Helium Foundation, for supporting open

LoRaWAN infrastructure worldwide.

482

Check them out at helium.foundation

and if the show has been

483

useful, a quick rating or review

on Apple Podcasts or wherever you listen.

484

This really helps

485

people find it and helps the show grow

so we can help more people.

486

I'm Nik Hawks with MeteoScientific.

487

I'll catch you on the next episode.