CC1190 bias current

- Do you know if your software controlling the CC1101+CC1190 is ok?

- LNA=1 is receive, as I read your post you want to use the CC1101+CC1190 as sender

- Which bias current are you referring to? Looking at the total current consumption for your board will give more information.

- If you look at the RSSI in the receiver, does this increase when you send with your prototype?

- 4.1 vs 3.9 should have very limited impact. The change is nearly within the tolerances on the components.

Thanks TER for your response.

I want the CC1190 primarily for its LNA, not really for sending. My application already works for the CC430 without the CC1190, where it stays in reception mode and passes the packets through the UART. I wanted the CC1190 to provide some extra sensibility but it's like if it were dead.

Could you post a schematic?

If you send a CW with max output power, is the RSSI on the receive side on the CC430+CC1190 minimum or does it go up when you turn on the Tx side?

I have a standard node with no CC1190 IC and two of my prototypes with CC1190 at 15 meters from a standard receiver (no CC1190), with a couple of walls in the middle.

When transmitting from the standard node, RSSI takes values around -5dBM

When transmitting from the first prototype with CC1190, RSSI takes values around -16dBM

When transmitting from the second prototype with CC1190, RSSI takes values around -40dBM

I'll post the schematics later this morning.

Thanks a lot TER.

And here is my latest schemmatics. This is a minimalist circuit. After cloning the reference design without success I decided to remove the SAW filter and hook the CC1190 directly to one of my working radios. I know this is far from being CE compliant but at this point I only want to make the circuit work.

L12 has been replaced by a 3.9 nH inductor as I said in my first post.

Thanks again TER for your time

.

This schematic is correct. I assume you have connected the PA_IN/LNA_OUT to a 50 ohm point in the CC430 part.

Have you confirmed that the control pins on CC1190 have the correct values?

It could be something wrong with the layout.

Have you checked the RSSI as I asked about?

Sorry, overlooked the post with the RSSI. Do you have a signal generator?

Since you are most interested in Rx, use the +CC1190 as receiver with a known good sender first. Simplify the test setup by having sender and receiver fairly close but without antennas, the walls just add a complication.

Thanks TER for your new reply.

TER said:

This schematic is correct. I assume you have connected the PA_IN/LNA_OUT to a 50 ohm point in the CC430 part.

Yes.

TER said:

Have you confirmed that the control pins on CC1190 have the correct values?

Yes.

TER said:

It could be something wrong with the layout.

Maybe but I've measured continuity and voltage levels all around dozens of times. I've also replaced the CC1190 IC in one of the latest prototypes in case that the reflow temperature were too high. I remember having burned several CC1101 IC's in the past.

TER said:

Do you have a signal generator?

Yes but it's limited to 40 MHz if I remember well.

TER said:

Since you are most interested in Rx, use the +CC1190 as receiver with a known good sender first. Simplify the test setup by having sender and receiver fairly close but without antennas, the walls just add a complication.

- Could you post gerbers of your layout (including the drill layer and the stack-up information)

- Have you double checked for assembly errors? I have seen before that strange errors are caused by one or more components mounted with wrong value/ wrong type.

- I find it strange if you have received bad CC1190s. It would be interesting if you are able to test the parts on a known good CC1190 design

TER said:

- Could you post gerbers of your layout (including the drill layer and the stack-up information)

Gerbers and drill file:

5381.production.zip

TER said:

- Have you double checked for assembly errors? I have seen before that strange errors are caused by one or more components mounted with wrong value/ wrong type.

Yes, many times but I should probably build some prototypes more and maybe order the parts from another supplier. This is in case that any of the parts were incorrectly labeled.

TER said:

- I find it strange if you have received bad CC1190s. It would be interesting if you are able to test the parts on a known good CC1190 design

I'm in fact looking for an eval board without the CC1101 IC since this would let me use it with my own radios.

Thanks again TER for your time.

Your layout is not following the ref design.

- Our ref design uses 4 layers.

- CC1190 is very poorly grounded, 9 vias should be used.

- Component placement not ideal.

- Vdd_pa1/ Vdd_pa2 should have a series component, see ref design.