Martin Harris - Moderator Posted March 12 Author Share Posted March 12 Weird, isn't it! I'll put another one together tomorrow with my "spare" bits. Could there be any difference connecting it via the spare pin method for direct mounting of the display on the Arduino - which way did you connect on test? Update - re the freezing problem, there does seem to be an environmental factor. In my living room, the scanner is showing a fair amount of 2.4 GHz activity (router and Bluetooth devices close by) and seems to run for extended periods but moving into my kitchen, where most of my testing has been done, signal levels drop dramatically and within a few seconds, the display freezes. However, the boot problem was occuring in the living room before I added the pull up resistors. Quote Link to comment Share on other sites More sharing options...
FlyinFlynn Posted March 12 Share Posted March 12 Very curious! Have you tried a different 9v battery?. Is it a primary 9v sort ie a duracell or similar or is it one of these li-ion 9v batteries? Perhaps, if it is one of the li-ion types, you have one which is actually 7.4v dressed up as a 9v with a dodgy boost regulator that is electrically noisy. 🤷♂️ I think the microwaves magnetron operates on 2.4GHz...... Quote Link to comment Share on other sites More sharing options...
Phil Green Posted March 12 Share Posted March 12 8 hours ago, Martin Harris - Moderator said: Could there be any difference connecting it via the spare pin method for direct mounting of the display I dont, mine have all been just the simple 4 wires, gnd, vcc, A4 & A5, with PP3 power via Vin and the 'other' gnd... which 'spare' pin? Otherwise as Mike says it has to be a power problem - maybe try some extra decoupling caps on the 5v? The EBL lithium PP3s dont generally have a boost reg, they're straight 2S, I use them a lot 🙂 Quote Link to comment Share on other sites More sharing options...
Martin Harris - Moderator Posted March 12 Author Share Posted March 12 I was referring to the direct board to board mounting method which I believe uses A6 linked to 0V and A7 bridged to 5V to feed power to the OLED. I’m using fresh Procell (industrial Duracell) PP3s and seeing a steady 5V at the 5V pin. On the first attempt, I established that it would work on any input voltage between 3.9 and 6 volts. Quote Link to comment Share on other sites More sharing options...
Phil Green Posted March 12 Share Posted March 12 Odd. I've no answers Martin, just thinking aloud, the only components that see 9v are the decoupling cap and the 5v regulator, the 328P part sees 5v regardless of input voltage and the NRF part sees 3.3v again regardless of input voltage. So the only components even aware of the battery voltage are the decoupling cap and the 5v regulator. I dunno, you have a known-good RF-Nano and a brand-new display - the only two components. So all that remains is the wiring and the battery. It runs for a while then stops? any warm components when it stops? Wires too long? since thats solid core wire not stranded could there be a break somewhere mid-run? How does it do on a 2S lipo? or any battery other than the Procell? Baffled! 😆 Quote Link to comment Share on other sites More sharing options...
Martin Harris - Moderator Posted March 12 Author Share Posted March 12 I don’t know if it helps but I knocked up a quick test assembly with my original Arduino and the new display. Guess what? It behaved almost the same and I’ve tried a Duracell on both too. I’m sure it’s wired as shown on the mode-zero site but in case I’ve suddenly developed colour blindness or electronic dyslexia, I’ve attached a photo of the test version and a video of it failing to initialise. A 10k pull up resistor to VDD got it further through the sequence but I didn’t bother trying to find the “magic” value(s) for this one. IMG_3997.mov P.S. I know the Duracell is out of date but it was still sealed... Quote Link to comment Share on other sites More sharing options...
Martin Harris - Moderator Posted March 12 Author Share Posted March 12 1 hour ago, Phil Green said: Odd. I've no answers Martin, just thinking aloud, the only components that see 9v are the decoupling cap and the 5v regulator, the 328P part sees 5v regardless of input voltage and the NRF part sees 3.3v again regardless of input voltage. So the only components even aware of the battery voltage are the decoupling cap and the 5v regulator. I dunno, you have a known-good RF-Nano and a brand-new display - the only two components. So all that remains is the wiring and the battery. It runs for a while then stops? any warm components when it stops? Wires too long? since thats solid core wire not stranded could there be a break somewhere mid-run? How does it do on a 2S lipo? or any battery other than the Procell? Baffled! 😆 Maybe a solution now but not really an explanation. It doesn't complete the boot sequence using a 2S Lipo - or a 2S LiFe for that matter but it does on a 1S Lipo with and without pull up resistors! I've left the new test assembly running in the kitchen for several minutes and it seems happy. I can see a couple of ways forward now but it does seem so odd that I've had these problems with the standard 9V power source. There seem to be two issues here - but are they related? First is the boot up - not a problem for anyone else and seemingly relating to the absence of pull up resistors to SCK and/or SDA. Second is freezing which seems to happen mostly, but not exclusively, in the kitchen where I'm seeing much reduced signal amplitude displayed when operating on the PP3. The only common factor which hasn't been eliminated - but one that doesn't seem to be considered a possible source of the problems - is that the display that I sourced from a separate supplier is physically identical to the first one. Worse in the kitchen...no, the microwave hasn't been in use while I've been testing but that does raise a separate concern...has anyone seen activity from one registering on the scanner or is mine leaking? Quote Link to comment Share on other sites More sharing options...
Phil Green Posted March 12 Share Posted March 12 (edited) Seems I was fibbing (in one case) - I just dismantled an older one and it seems I did add 4k7 pullups to this one after all. But generally I dont. Honest! 😀 Edited March 12 by Phil Green Quote Link to comment Share on other sites More sharing options...
Phil Green Posted March 12 Share Posted March 12 Forgot to say, the value isnt at all critical, 1k to 10k ish, its just that I2C is open collector and the pull-ups return the inactive sda & sck signals to their rest state (logic high). Its so the I2C lines can be driven by multiple devices. Quote Link to comment Share on other sites More sharing options...
Martin Harris - Moderator Posted March 12 Author Share Posted March 12 I’ve gone with a vape battery - luckily my daughter is staying with us this week and had just finished one - what a waste, lasts a week or so but unrefillable apparently but with a 650 mAh LiPo cell and neat little USB C charger module, complete with LED charge state display. Unfortunately, I damaged the display during the excavation of the module which seemed to disable its charging function so went with a simple charge jack. While 2S would probably be better, it seems happy to work for several hours while maintaining sufficient voltage to function effectively and there are no balance charging considerations (although my ISDT charger gets a little excited by the prospect of doing an unbalanced charge!) I may modify the case design to incorporate a charger module and display once I’ve worked out its switching arrangement (some sort of suction operated switch in the discarded parts of the vape, but I don’t recall any wiring) - I’ve got my name down on the next vape! Quote Link to comment Share on other sites More sharing options...
FlyinFlynn Posted March 13 Share Posted March 13 19 hours ago, Martin Harris - Moderator said: Worse in the kitchen...no, the microwave hasn't been in use while I've been testing but that does raise a separate concern...has anyone seen activity from one registering on the scanner or is mine leaking? Thats funny.. I posted a link to a short youtube video I made of my microwave leaking but it doesn't appear to be showing up here yet 10 views have been made???? Where did I post it I wonder? Here is the link youtube video Quote Link to comment Share on other sites More sharing options...
RottenRow Posted March 13 Share Posted March 13 Martin, have you tried running these with a 9V power supply rather than the PP3 batteries? Also feed the 9V straight to the board, missing out the battery clip. It’s unlikely to be the clip causing the issue but it’s easily eliminated. I had problems in the past with the Procell 9V batteries, poor performance and leak when they get old. Their AAs were fine but not the PP3s or whatever they’re now called. Also, have you tried linking together both GND pins on the Nano? Worth trying for the time it will take. Brian. Quote Link to comment Share on other sites More sharing options...
Martin Harris - Moderator Posted March 13 Author Share Posted March 13 Thanks for the ideas Brian. I established an input range of 3.9 - 6V which would work using an adjustable power supply connected direct to the board so I can discount battery or connector issues. Funnily enough, I did wonder about the ground connection and tried a link across the board but that made no difference. It’s so strange that I’ve had these problems and I’m perfectly prepared to find that I’ve done something incorrectly but I’m at a loss to see what that could be with such simple circuit connections. I’ve built two examples (both from the instructions, not copied) and they both show exactly the same characteristics, despite one Arduino having been programmed and tested by Phil. The only unknown is the OLEDs - which were sourced separately but physically identical although those more informed than me can’t see any way it might affect these symptoms. It would be interesting to discover whether anyone else has built one from currently available components and how they got on. I believe 99% of us are using Chinese clones of Arduinos so there might be some differences from older versions? FF - thanks, it seems like the shielding on microwaves isn’t totally effective then… 1 Quote Link to comment Share on other sites More sharing options...
Brian Dorricott 1 Posted March 13 Share Posted March 13 Just looked at this thread , where can I find the build instructions and code . Not sure it's an existing topic or not searching didn't help. Quote Link to comment Share on other sites More sharing options...
Phil Green Posted March 13 Share Posted March 13 (edited) Here you go: https://mode-zero.uk/viewtopic.php?f=42&t=1008 Be aware that there are several variations of the RF-Nano, we recommend the "Keywish" one. Otherwise a separate NRF24L01 and Pro-mini or Nano works exactly the same. I should add that within weeks of publishing, Martins design was ripped off by Lantian, a Chinese manufacturer - if you can find one, their complete unit is probably cheaper than the parts, time & effort: https://www.himodel.com/m/building_tools/LANTIAN_Portable_High_Precision_2.4G_Spectrum_Analyzer_OLED_Display.html Lantian didnt even credit Martin, all they did was to change the splash screen from 'Ceptimus' to 'Lantian'. This is really bad etiquette. HK did the same with my dual-receiver adapter 😐 Cheers Phil Edited March 13 by Phil Green Quote Link to comment Share on other sites More sharing options...
FlyinFlynn Posted March 14 Share Posted March 14 19 hours ago, Phil Green said: I should add that within weeks of publishing, Martins design was ripped off by Lantian, a Chinese manufacturer - if you can find one, their complete unit is probably cheaper than the parts, time & effort: Where is the fun in that though Phil? 22 hours ago, Martin Harris - Moderator said: FF - thanks, it seems like the shielding on microwaves isn’t totally effective then… Considering the microwave was pumping out 800watts and the scanner showed only a very low signal (sort off) I don't think the shielding was doing a bad job... about the same as standing in front of your wifi router or having your mobile pressed up against the side of you head for hours at a time ( not that anybody on this illustrious forum would be doing such a thing!)🙂 Quote Link to comment Share on other sites More sharing options...
Martin Harris - Moderator Posted Wednesday at 22:37 Author Share Posted Wednesday at 22:37 n an earlier post, I mentioned that I might do a redesign of the case to incorporate the charging module/display from the disposable vape. I’ve finished my “production” version which seems to function well on the single vape battery (650 mAh LiPo) and now shows the battery state for about 10 seconds after power on (and about 4 seconds after switching off). An evening’s continuous use reduced the state to 90% so even if it isn’t happy at lower voltages, there’s a decent usage time (not yet established) and it recharges quickly (and can be used during charging) from any USB source. Displays after powering on The internals The vape to look for! Power supply connection details If anyone wants to copy this, the new STL files are attached. Case.stl Base.stl Cell Support.stl Bezel.stl 2 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.