开云体育

The value of the device in the schematic diagram does not display the real value. Please pay attention to the text

Re,

If I look at the schematic, it’s four identical filters.
Where you saw that: "the schematic diagram notes that TG uses 470MHz." ??


cdt

Do you have a 470MHz SAW? I see that the schematic diagram notes that TG uses 470MHz, and I saw it at first, but later forgot it, which led me to weld 433.92 B3555

I see that the last page of the datasheet has a clear description of the welding temperature


The above is the temperature of the device. The actual welding equipment temperature must be slightly higher to make the device temperature reach the target value.

I guess you don’t have solder paste for patching(If you have solder paste, the operation is the simplest. I use solder paste when welding). If I am in this condition,
1.I will use a soldering iron to put a small amount of solder on the pad on the front of the PCB;
2.And then blow it with a heat gun(hot air of JCD 8898) on the front to melt it;
3.Use the fastest Switch to bottom heating at high temperature, observe the color of the tin, if it lights up, it means that it is in a melting state.
4.At this time, quickly put in the B3555, and the melted tin will automatically correct the position of the device.
This operation is a test of hands-on ability.

If you still have flux, soldering should be easier. After tinning the pad(step 1 above), use hot air to keep the pad heated, and at the same time apply flux to the B3555. After the solder melts, apply the flux Put the B3555 into the specified position, so that the welding will be more reliable and safe.

Brilliant!? So now you have your "fun of DIY" mood back!

Well done.

I have dissected a dead mouse this evening, extracted the encoder, connected it up to my old prototype board and experienced the same issue.? I also tried a really basic encoder that also behaved in the same way, so I think it is the encoder I am using that is unusual.

I think your idea of using a mouse encoder and a 3d printed wheel is really good.

Have a great weekend

Before I selected the encoder, I read the datasheet of EVQWK that you designed and selected. It is a mechanical encoder that provides quadrature pulses. Therefore, I selected the mouse encoder according to this specification. The model I selected belongs to a similar mechanical encoder.



I think when you use EVQ-WK, its effect should be to scroll a joint to trigger a scroll signal, which is to scroll a line of menu options, but unfortunately, the effect of using the mouse encoder is not the same.

I read your program code, there is software filtering in it, so there is no additional capacitor added to the hardware, I also tried adding this capacitor, but the effect did not improve.

When I used the disassembled mouse encoder, I cut off all the three pin leads of the encoder. My ESP32 board has pull-up resistance according to your design.



And,I?change bool fullQuad=false to bool fullQuad = true did not succeed.




The original intention of changing this mouse encoder is because the mouse encoder is more common and easier to get than EVQ-WK, but the current state disappoints me.

The number of pulses per revolution should not matter.

?

The encoder I use has a mechanical click for each step, one complete 360 revolution would result in 15 steps.? Think of the scroll wheel on your mouse – it has mechanical clicks (at least all the ones I have used do).?

The EVQWK encoder provides quadrature pulses.

?

See this link for an explanation.

?

It looks as though your encoder is also quadrature however, but maybe optical?? Is the encoder connected to some other IC and the output from that used to signal to the simpleSA?? It could be that IC is increasing the resolution.

?

It may be there is some “bounce” on the contacts (multiple changes of state) that could confuse.? Maybe add some capacitance to try and smooth out the pulses, though should not really be needed.

?

Also note that these particular input pins on the ESP32 do not have an internal pull up/down resistor – have you included one?

Another option to try would be in encoderSA.h line 58 change bool fullQuad=false to bool fullQuad = true

Dave

On Mon, Jan 2, 2023 at 07:59 AM, Gerard wrote:

I received the PCB's from JLCPCB. Again the PCB's are superb, nice work. These are the most small PCB's? I’ve ever made by JLCPCB. I’ve had them for about ten days and I always wonder how to weld it with an iron. Being a bit ill I have not yet started, but it will not be long.
Whaooow is small. LOL

On Tue, Jan 3, 2023 at 02:20 AM, m0wid wrote:

使用示波器和低电容有源探头检查进入 SI4432 的波形。我想最坏的情况是 SI4432 已被初始错误连接损坏。可能更有可能是分压器电容值需要一些调整

Finally succeeded, the problem lies in the PIN1 pin of the active crystal oscillator, the datasheet is defined as Tri-state, the test screenshot has a switch for grounding, I have encountered it before, but I can’t remember whether the grounding is on or off, so I connected a resistor R15, it turns out that this is correct. After I removed R15, the REF output was normal, and there was no need to calibrate the frequency. Haha, the mood at the moment is the fun of DIY.
It's just a small step though, there are other features to check.




Seeing the Reverse Beacon Network you described, I learned about it, and I feel that it is very difficult. Is there any other PC or mobile terminal software?

I see you have found Si Labs AN417 describing the requirements for external oscillator, which I had not looked at before.
I think the best approach would to use an oscilloscope and a low capacitance active probe to check the waveform into the SI4432, after checking the signal from the NOT gates.? I guess worst case is that the SI4432 has been damaged by the initial incorrect connection.? Probably more likely is the divider capacitor value needs some adjustment.

C103/R85/89 in your oscillator circuit - are they needed?? If C103 is kept maybe apply some DC mid point bias to pin 1 of the NOT gate.

What frequencies do you see from the other SI4432?

Dave

Oops - the settings.html, help.html and backupRestore.html files are a hang over from another completely different project (I used some of it as the base for some of the wifi code) and can/should be deleted.

If you are interested the Reverse Beacon Network can be seen here:?

Dave

Hello,

I received the PCB's from JLCPCB. Again the PCB's are superb, nice work. These are the most small PCB's? I’ve ever made by JLCPCB. I’ve had them for about ten days and I always wonder how to weld it with an iron. Being a bit ill I have not yet started, but it will not be long.
Whaooow is small. LOL
cdt

I don't understand what the Reverse Beacon Network is?

On Sat, Dec 31, 2022 at 02:22 PM, m0wid wrote:

Looks great! ?Not using the modules allows much more layout flexibility. Did you find a 30MHz TCXO? ?Looking forward to hearing how it performs. Nice you have an oven for soldering, that attenuator was tricky by hand.?

What encoder do you plan to use?

Will you share this design?

Thanks

?

David Wilde

Of course, if you are interested, I will check my design and share it after confirming that there are no obvious mistakes; I do not use TCXO. TXCO is not easy to buy. I checked that LCSCs are out of stock. At present, they are replaced by active crystals with product number C2831390 and model OT322530MJBA4SL. Its advantage is that its packaging is similar to TCXO. In the future, you can consider whether to buy TCXO based on the actual performance; Encoder I plan to use a mouse encoder. The scroll wheel is 3D printed to meet the size requirements, or you can use the mouse's own scroll wheel and add a touch button, which can achieve the same functions as EVQWK



Now I'm trying to compile the program and try to light up the board