nrf52832 ble problem

RE: nrf52832 ble problem

Navdeep — Tue, 03 Jul 2018 20:28:34 GMT

I think so because its one of the best companies for component placement here in canada and I'm working on the connections , will let you know when the board works

RE: nrf52832 ble problem

AmbystomaLabs — Tue, 03 Jul 2018 20:12:51 GMT

So out of curiosity, I'm guessing you didn't populate this board?

And, did the person who populated it with components do correct reflow work?

RE: nrf52832 ble problem

AmbystomaLabs — Tue, 03 Jul 2018 20:04:23 GMT

And finally here it is with the excess wire snipped off and an nRF52832 placed alongside for size comparison.

RE: nrf52832 ble problem

AmbystomaLabs — Tue, 03 Jul 2018 19:56:29 GMT

No, no, no, that won't work at all. The reason the ground problem is bad is due to the physical length of the ground connection. Ideally we are looking for about 1mm of wire to bridge your ground. What you did will not work.

See below for the correct way to do this. A trace on the 52840 DK was bypassed for to demonstrate this technique.

RE: nrf52832 ble problem

Navdeep — Tue, 03 Jul 2018 18:39:13 GMT

HI please find the attachments. I think because of small size, I am not able to connect wires in a proper way. May be, that's the issue

RE: nrf52832 ble problem

AmbystomaLabs — Tue, 03 Jul 2018 16:57:49 GMT

You should do 4 minimum on the ground plane side as I showed in the drawing. It really should start working. Let me know. It shouldn't take you more than an hour to accomplish and test. Time well spent.

Also, please post of photo of your re-work when you reply.

RE: nrf52832 ble problem

Navdeep — Tue, 03 Jul 2018 16:53:33 GMT

I have done 2 connections with ground to the nrf pad (49) by scratching off the solder mask one connection through the cap c7 and c8 via , but still the problem persists. I think this is the only issue. SO should I go for new layout ? I will mail you the new design for the review.

RE: nrf52832 ble problem

FormerMember — Mon, 02 Jul 2018 11:26:01 GMT

(It is also possible to upload layout for review in private case here on devzone.)

Thanks @AmbystomaLabs for such good and detailed answers!

RE: nrf52832 ble problem

AmbystomaLabs — Sun, 01 Jul 2018 13:23:27 GMT

You can send your updated layout for review to my email if you wish.

Yes the local capacitors are not doing anything because the length of board surface the current must flow is so long that there is significant resistive losses and parasitic inductance. Also since the ground currents for the entire nRF pass through that same narrow connection they talk to each other ( or couple). These are often called ground loops and are a large source of noise in a device.

RE: nrf52832 ble problem

Navdeep — Sun, 01 Jul 2018 01:57:18 GMT

ohhh.. I see.. Thats a big mistake. The ground Pad of the nrf is connected through gnd just through cap C8 and C7 which is not sufficient, Thanku so much for pointing this out. I think this is the only problem.I will try to do the connections manually and for my new design checking that I will make, can I send the layout on your email id (richard.moldovan@ambystomalabs.com) or should I post it here ?

and more thing. You said this

"The way you have done the layout on your board none of the local capacitances are really doing anything since they are not connect to the same ground as the nRF."'

because the nrf pad gnd is not at all connected to the gnd, its just through the c8 and c7. right?

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 22:13:28 GMT

Not sure why your cad app didn't pick up the mistake. But the connecting dot at pin 49 is suspicious. It's possible it didn't think the ground slug on the nRF was part of ground.

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 21:58:33 GMT

You don't have to do the connections precisely as I have shown, but you need at least 4 good connections around the chip to solve the problem and more connections is better.

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 21:55:57 GMT

Well broken ground plane seems to be your problem.

As I show in the following image you have essentially isolated the ground of the nRF from the rest of the circuitry. This is very, very bad. It looks like the nRF might have a sliver of a ground connection at C7, C8 but even that is marginal.

The way you have done the layout on your board none of the local capacitances are really doing anything since they are not connect to the same ground as the nRF.

As to why it works when you hook up the gpio, some level of physics magic is happening. Perhaps you are stabilizing the ground reference with that connection.

For starters to prove to yourself that is the problem, solder some 30 gauge wires across your broken bottom layer in a few places and the problem should go away. But you are going to have to redo your layout to fix it.

I recommend you pass your updated layout by me for review when it is ready.

Bridge the ground like this and see if it starts working. Just scratch off some of the solder mask to connect the wires. Use really small gauge hookup wire, like 30 or 32 even 40 gauge is fine. On the Vss connection at pin 45 you can probably bridge it with a little dab of solder. Just be careful not to short the pins. If this connection scares you, then do the others first and see how it goes.

RE: nrf52832 ble problem

Navdeep — Sat, 30 Jun 2018 21:12:28 GMT

Thanku so much, I will go through the above link. and please find the following attachments-

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 20:48:36 GMT

Please post the ground plane layer too.

On your question regarding the 0805 caps, coin cell batteries have very poor current sourcing capabilities. In my lab, I simulate them by putting 200ohms in series with the bench supply.

I talk about it a lot on this site. You should read the Texas Instruments white paper on the subject and also look at a coin cell spec.

Using a coin cell and no external current consuming circuitry, the best you can expect is about 2% duty cycle assuming you put about 22uF of MLC caps on the board. Thus why I use 0805 parts. Our design actually has an external PA uses much more power. So we place about 100uF of MLC caps. 2x 47uF 0805 and 2x 2.2uF 0402.

Here are some recent devzone comments on it: https://devzone.nordicsemi.com/search?q=ambystomalabs%20ti%20coin%20cell

RE: nrf52832 ble problem

Navdeep — Sat, 30 Jun 2018 19:51:29 GMT

Hi, Please find the below attachment. I would really appreciate if you can sort out the issue.

In conclusion, we have to short any pin( P0.07 to P0.10) with 3.3v , and then nrf start advertising ,not without that. SO please check the attachments and let me know where we are doing wrong.

Also, as you have mentioned that your product is cell based, and for that you are using 0805 package. In our case, we are also going to give power through button cell ( but now, we are just powering it through constant 3v supply from another board), so please explain me properly the significance of 0805 in this case.

I would really appreciate.

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 14:28:22 GMT

I should have said “capacitance value”. For the logic portion of the chip, the physical size is not too important. But since the SRF ( series (or self) resonant frequency ) is strongly correlated with the value, it is best to use a combination of large and small value caps as they do in the reference design.

On the RF the physical size of the cap influences the size of the pads and microstrips that should be used. This is more a method of controlling parasitic reactances on RF.

Our own designs make use of 0201 parts for much of the logic and all of the RF. Only bulk storage capacitance is made up of 0402 and 0805s. We utilize 0805s since our main product is a coin cell powered device. Thus we need a lot of local capacitance.

happy to review your design and provide feedback.

RE: nrf52832 ble problem

Navdeep — Sat, 30 Jun 2018 13:49:57 GMT

Ohk I will post you the schematic and layout soon. But a quick question,as you have said, the capacitor size matters!!.

I have used various 0201 capacitors instead of 0402 and 0603 but with the same specs given in nrf datasheet. Does that honna make any difference?

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 13:43:44 GMT

Much more clear. Thank you.

I would surmise that you failed to connect one portion of the nRF's power structure to the rest of the power structure. There are 3 x Vdd connections and 3 x Vss connections. All must be hooked up properly and with the sufficient local capacitance. Likely either the connections were broken in your layout, or the ground is broken with ground loops forming, or capacitances are wrong size or too far from the chip to be of any use.

If you wish, please post your layout and schematic for review. If you consider it to be confidential, then follow the Nordic process to have their team review your design via a private posting.

RE: nrf52832 ble problem

Navdeep — Sat, 30 Jun 2018 13:32:53 GMT

My point is, I have already powered the microcontroller through gnd and vcc pins. I dont want the controller to source power through gpio pin. Controller board functions are working fine without connecting that gpio pin to vcc. The problem is it advertise only when i connect that gpio to 3.3 ( still at that time ,its powered through power pins as well)

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 13:28:54 GMT

Perhaps you should try re-reading my answer. I did not mention ESD protection guidelines.

What you described in your question is the phenomena of the chip sourcing power through its gpio. You said you hook up power to P0.07 and the radio starts advertising. This is normal, as all microcontrollers will attempt to source power through their gpio.

If there is a different problem you need help with, try rephrasing it since it was not clear in your question.

RE: nrf52832 ble problem

Navdeep — Sat, 30 Jun 2018 02:17:55 GMT

I have considered proper esd protection guidelines while designing layout. What can I do to sort out this thing?

RE: nrf52832 ble problem

AmbystomaLabs — Sat, 30 Jun 2018 00:24:43 GMT

It is normal for microcontrollers to source power through their gpio. Both the FET and ESD diode structure contribute to this. For this reason all gpio connected devices need to see the same voltage levels either by running them all on the same supply or using a combination of level shifters and/or open drain devices with pull ups.