Lag with 202 protocol but not others with nrf24l01

Hi,
I'm new to this, and have a problem with lag.
I have just installed a nrf24l01 (one from ebay but with the lna and pa) and am pretty confident about the soldering. I have 4 quads which need this module:
Cheerson cx10 (new version)
JJRC H8 mini
JJRC jj-1000
Tarantula X6

The last 2 use the 202 protocol and suffer from control lag that makes them un-flyable. I can bind no problem but there is roughly a 0.5 to 1.0 second lag in the response. The cx10 and H8 both fly perfectly so I don't know what to think.
I used a ribbon cable to attach the module and I wonder if there is a problem with the current capacity of the thin wires.
I've flashed deviation 4.01 and the most recent nightly 4.01-1194044 and the symptoms are the same.

Any help much appreciated.
thanks.
DrT

There could be many things in play, and many have reported success with these models. So start out be using a very simple model.ini... Set up a new model with the four basic controls -- when I am unsure of a file I usually remove the blades -- and see if the problem still exists. If it does, it may be a hardware problem, and you don't need to mess with software until you rule that out.

For hardware issues we need to see a photo of the installation with enough detail to see the circuits. Let us know what you find out.

The 202 protocol looks for packet acknowledgements from the aircraft. The protocols for the other quads you listed do not.

There's a thread somewhere on here about some knockoff nRF chips having a bit reversal bug that affects receiving packets. It could also be a wiring problem on the MISO signal in your nRF install.

Thanks for the replies.
Here's pictures of my soldering and the module.

Soldering

Soldering

Soldering

Soldering

Soldering. The last connection looks a little dodgy but it's good underneath.

Thats no soldering thats baking!

Soldering looks ok, I think. The only thing I noticed is that there might be a hairline wire between the yellow (CSN) and orange (MOSI) wires on the module end. But a short there would probably cause problems whenever you fly, so I may just be misinterpreting the photo. It appears that everything is wired to the correct points.

It's possible the quads you bought came with a different protocol or even chip than was used to create the protocol we use now. Did you have any luck with a barebones model.ini?

It's not a hairline wire, the board has some sort of lacquer on it, and the line is where the flow is uneven.
The quads are pretty common, I can't see both being so different that nobody else has experienced this.
My feeling is that the nrf24l01 is to blame. Looking at pictures online, it looks quite likely to be fake. Although this doesn't necessarily mean that it doesn't function the same as an original, I've now read a few reports of problems that might be relevant.
I don't know how to start with a barebones model file. Can you help with a link to a tutorial or explain the basics?

thanks,
DrT.

It has been suggested to buy the nrf24 from banggood, just to be sure. The worst part of that is waiting for it to arrive.

It seems crazy that I might have a clone from a UK supplier and the solution is to order one from China! However, that is exactly what I have done.
I'll report back with the results when I've received it.

DrT

So my module from Banggood arrived today. I ordered the one listed as "DEVO Transmitter Tuner Modified NRF24L01". It's about two thirds the size of the one I'd used previously.
I soldered it in and everything works perfectly, no lag!
I'm now more inclined to think the other one was a fake/clone.
Thanks everyone for the advice.
DrT.