HC-12 working supply

Grogster

Senior Member
@ reginald: The plot thickens, and the crystal would seem to be our problem child. I would consider replacing the crystal and using the clones, but I am not really that comfortable trying to force cloned modules into working, as the silicon used for the MCU and RF chip are probably also cloned, and could be unstable. Thank you very much for performing that operation. :) That was something I was thinking of doing for the purposes of experimentation myself, and now I don't need to. :)
 

PhilHornby

Senior Member
Crystal measurements...

Here is an initial comparison of the signal across a crystal marked T300, vs one of Grogster's (marked E300).

CrystalComparisons.jpg

I'm away from home for a week - but will do some more probing on my return....
 

Grogster

Senior Member
WOW!!!!!!!

.....stunned silence........

That would explain quite a lot, really. That "30,000MHZ" crystal is freakin' MILES out.
This is still very good information to have though, so my thanks to Phill for taking the time to measure my clones. :)
I await your further testing when you get back with much anticipation!
 

manuka

Senior Member
Indeed stunning ...

With a view to alerting the wider e-community I've taken the liberty of a "mailbox" message to the Australian "Silicon Chip" monthly -
SiChip: This may suit publication ?

Clones of the cheap,popular and versatile Chinese HC-12 UHF FSK data transceivers have shown problems when used together with earlier HC-12 modules. Compatibility issues seem to relate to clones "30.000 MHz" crystal producing 400 kHz spaced channels that are 37kHz above the correct frequencies. In contrast,original HC-12 modules using "T300" crystals very closely match stated channel specifications (433.400 MHz, 433.800 Mhz, 434.200 MHz etc ) .

Mix and matching classic and clone modules should hence be avoided, as clone off spec. transmissions may fall outside receiver pass bands. These HC-12 clones seem to otherwise work OK,although signals at some channels may fall beyond legal band limits.

Identification seems best made from a close inspection of the module's canned crystal. Maker markings may also be present,along with subtle silk screening & masking differences. A spectrum analyser or UHF scanner could further show the actual transmitter frequencies.

View thread => http://www.thebackshed.com/forum/forum_posts.asp?TID=10443&PN=1&TPN=1

Regards - "Manuka" (Stan. SWAN - Wellington,NZ)
 
Last edited:

AllyCat

Senior Member
Hi,

That "30,000MHZ" crystal is freakin' MILES out.
Can it really be that far out? I thought we were looking for a discrepancy of about 2.6 kHz (@ 30 MHz).

I can't see any horizontal scale calibrator on the 'scope waveforms but both are close to 13.5 cycles across. I would have expected 28.5 MHz to be about half a cycle less. :confused:

It's a long time since I worked specifically with crystal oscillators, but their frequency was often rather more "circuit-dependent" than (I) expected or desired. :(

Cheers, Alan.
 

techElder

Well-known member
My daddy used to tell me, "When you set a fire in the barn, all the horses are going to want to get out whether the barn is on fire or not."
 

premelec

Senior Member
Bending xtal frequency is still done for small range VFOs - inductors or more commonly capacitors are used... what's not clear in this situation is just what the schematic of the device is and what might be possible to alter [like capacitors].
 

reginald

New Member
I had three HC-12 30mhz crystal variant left whch did not communicate well(or at all) over distance with my T300 type. I obtained new crystals from farnell type

7M-30.000MEEQ-T - Crystal, 30 MHz, SMD, 3.2mm x 2.5mm, 10 ppm, 10 pF, 10 ppm, 7M Series stock No 2308724

Swapped these today and now all three boards are happy chatting away with no errors. One was a USB type plugin which failed and on opening up it had a HC-12 board piggy backed on the USB board which I had swapped with 30mhz type (before I new the difference) and wondered why it didin't work. Shows not all crystals are equal.
 

Grogster

Senior Member
Great to know - thanks for sharing. :)
I have also ordered some crystals from Element-14, in the hope that this might fix MY crook ones, and now I am quite hopeful based on your post above. :)

I ordered some of these crystals, but they seem to be the exact same part - great minds think alike. ;)

Can you describe how difficult(or not) the procedure was? I expect quite painless armed with an SMD reflow station, yes?
 

techElder

Well-known member
Now, is it possible that the original manufacturer built a bunch of these HC-12s not imagining that the supplied crystals would end up being off frequency? Then provided them to eBay distributors really cheap, who gladly sold them off at discount prices? Isn't that a probable scenario?
 

Grogster

Senior Member
Sure. Anything is possible. It's more likely they are clones though IMHO. Anything that gets popular is cloned. The HC-12 website draws attention to the differences etc, which is more in line with a manufacturer trying to show you how to spot the non-genuine ones from genuine ones.
 

hippy

Technical Support
Staff member
Now, is it possible that the original manufacturer built a bunch of these HC-12s not imagining that the supplied crystals would end up being off frequency? Then provided them to eBay distributors really cheap, who gladly sold them off at discount prices? Isn't that a probable scenario?
It is possible but does not seem very probable. Doing that would risk damaging their own reputation and I don't think they would be quite so eager to label them as fake if they were involved in that.

I would tend to agree with Grogster that they are more likely clones with the manufacturers of those possibly not caring that they are not compatible with the genuine article or that their components may be out of spec to what should have been used.
 

techElder

Well-known member
Reason being that the so called clones in my inventory are exactly like the first ones I bought in every way I can find, except for the crystal label.
 

reginald

New Member
Great to know - thanks for sharing. :)
I have also ordered some crystals from Element-14, in the hope that this might fix MY crook ones, and now I am quite hopeful based on your post above. :)

I ordered some of these crystals, but they seem to be the exact same part - great minds think alike. ;)

Can you describe how difficult(or not) the procedure was? I expect quite painless armed with an SMD reflow station, yes?
I have got a station with a soldering Iron and hot air gun like in link below and it as worked well for a few years now

https://www.ebay.co.uk/itm/YiHUA-898D-SMD-ESD-SAFE-2-IN-1-HOT-AIR-REWORK-SOLDERING-IRON-STATION-LED-UK/162865657180?hash=item25eb8cb55c:g:g3YAAOSwbihaK1Yu

I placed a board on a heat proof surface and weighted it down with the end of a pair of pliers to stop it moving. This also acted as a heat sink on the processor next to the crystal. I set the heat gun to about 250C and a low air flow and using a small nozzle I pointed from the middle of the board to the outside edge over the xtal. I kept trying to lift it with a pair of tweezers and in about 10 secs it lifted off. In most cases there was enough solder left on the pads so no more was needed. Put a dab of flux on the board pads and placed the xtal with the writing the same way as with my original T300 boards. I didn't check to see if orientation made a difference. A bit more warm air as before( about 10-15secs) and the xtal floated in to place. Cooled it off, washed it with methanol to remove flux and all but one worked straight away. The dodgy one I removed and it had two dry solder pads, so a touch of solder on the board pads and reheat and it worked fine. I have swapped five xtal's now and they have all worked ok. Takes less than a minute to swap the xtal over. I think the only difference between the original and 'clone' boards is the xtal as my 'clone' boards seem fine after the swap. Perhaps the slight frequency difference shown between the two different xtal's is far enough apart so that you can't mix and match the boards at not least over a distance. I assume people using the 30mhz xtal boards have no problems like I have none sticking to T300 boards. Be interesting to see how Grogster you go on. The xtal's you have picked seem identical in specs to what I used (10ppm etc). Spent quite a few hours checking my software for errors until I found it was a hardware problem.
 

Attachments

Grogster

Senior Member
@ Texasclodhopper - Can I ask you if you have two different types of module? What I mean by that, is do you have ones with the crystal marked as T300 and also ones with the crystal marked as 30.000MHz or e300? The former has a nice thick white silkscreen on the bottom, the latter has a milky-white silkscreen that you can see through. The ones you have will work just fine, if they are ALL clones. If you have some modules marked as T300, and others marked as 30.000MHZ or e300, the latter won't talk happily with the former. SOME messages will get through, but most are lost. If they are ALL the same - no problem, as they are all off-frequency by the same amount, so work fine. I hope you see what I am getting at. My question really stems from my wondering if all you have are the clones. In which case, ALL of your modules will work fine together, as you are not mixing and matching the two different ones if you get my drift. :) What I can confidently say at this point, is that it is not just me dreaming. Phil and reginald have both confirmed the problem with their modules, and reginald has proved it 100% by his crystal swapping, so I believe my comments thus far are justified. :) I am still interested in your point of view though, so if you think there is still something I/we are missing in all of this, please do let me/us know what it is. All knowledge is power. :D

@ reginald - Thanks very much for all that extra info, and the pictures. I await my crystals with much anticipation. ;) I have an SMD reflow station, so should be able to follow your lead.
 

techElder

Well-known member
I'm old but I can follow. I've already outlined what I have. I'll just start a QC test on incoming modules. I'm not going to rebuild someone's rejects.
 
Top