Pico Prototyping - ...
 
Notifications
Clear all

Pico Prototyping - Building a "Pico Uno"

88 Posts
10 Users
15 Likes
5,768 Views
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 3 years ago
Posts: 6662
 

@tracecom So that's why it looks so much better than my pcb's, you are making a commercial product, you can afford to put more money into it than I can. What is it called?

First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, and 360, fairly knowledge in PC plus numerous MPU's and MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
Sure you can learn to be a programmer, it will take the same amount of time for me to learn to be a Doctor.


   
ReplyQuote
(@davee)
Member
Joined: 3 years ago
Posts: 1609
 

Hi all,

   I hadn't watched this video before, so it was only when my name got mentioned, that I caught up. I should say the only Pico I have is still in its bag, waiting patiently for its 15 minutes of stardom. This note (sermon?) is about 'general thoughts' ... many specifics will be omitted ... and sadly it has not been researched to the level of Bill's video.

As always, Bill (@dronebot) has done an excellent job. Not only is it technically informative ... I found it thereaupeutic listening to someone who has done their homework and knows what they are talking about .... thanks Bill!

As for power supply discussion that has been going on, I offer the following thoughts ...

  • Linear and buck (switch mode) power supply chips/modules both aim to do the same thing, and in suitable cases are interchangeable, PROVIDING the substituted unit is also within its ratings. (There are exceptions to this rule, but for the project discussed, I think the rule would usually be valid.)
  •  
  • Checking the ratings of the substituted unit meet the requirements of the application is not trivial, as a simple rating like maximum current =800 mA usually means exceeding 800 mA is inviting failure, but it does NOT mean it can cope with 800 mA under all conditions. It may only be able to pass (say) 400 mA or less if the input voltage is above a certain value. This particularly applies to linear regulators, but switch mode regulators can also have a lot of 'fine print' in their technical specifications which needs to be considered for a full relaible design. (Of course, in a low power hobby build, with few safety risks, you probably will not 100% analyse every aspect, but some basic checks are necessary to minimise disappointment.)
  •  
  • As already mentioned, buck regulators are more efficient at delivering power (except at low currents), so they tend to have much less problem with overheating.
  •  
  • So are the AM1117 and the small buck regulator interchangeable ... only in some cases!
  •  
  • The most obvious difference is power dissipation. I am including here a simple description of the thought process ... I am including sample numbers for illustration, but you should repeat the calculation using more precise numbers from the datasheet, etc. for your own project.
    • Say your external power source is 12V and the Pico needs 5V, so you consider the 5V AM1117 (or the variable voltage AM1117 with resistors set to give 5V) . The AM1117 drops the voltage by acting like a 'magic resistor'. In this case, the voltage drop is 7V (12V - 5V), which the data sheet says is acceptable.... but this is a conditional acceptable ... it is only acceptable if it doesn't imply breaking another rating!
    •  
    • If the current flow is 100mA, then the AM1117 must dissipate 7 V x 100 mA = 700 mW.
    •  
    • As the AM1117 has no heatsink, other than the PCB, this could easily be enough to cause it to overheat. (The heat dissipation capability depends upon lots of factors, including the PCB design .. see the datasheet for some guidance.) I haven't checked, but I think 700 mW is in the danger zone, which implies it might work, but problems are quite likely.  If that is the case, then we may extrapolate ...more than 100 mA demand will cause problems, but maybe 50 mA demand, which would reduce the dissipation to 350 mW, would be acceptable.
    • So could you use an AM1117 ... only if the 'sustained' current demand was below (say on my sample data) 50 mA.  Increasing it to 100 mA, might work, but the AM1117 is likely to overheat.
    • Note, the current is an 'average' current over seconds to minutes. A switch on peak would not normally be an issue as it will not cause appreciable temperature rises, unless it exceeded the maximum current rating of the device of (say) 800 mA.
    •  
  • I had a quick look at Pico current demand in https://datasheets.raspberrypi.com/pico/pico-datasheet.pdf
    • This suggested maximum currents of just under 100 mA, although presumably this is only the Pico itself .. sensors, LEDs, etc, powered by the board would increase this value
    •  
    • Overall, this suggests AM1117, with a 12 V supply, would be stretched to the limit and likely to be problematic, at least in the longer term. However, an almost 'bare' Pico board is likely to work, at least for a reasonable period of time, so AM1117s catastrophically failing almost straightaway might indicate other problems are also present. (Note from the Pico datasheet, current demand was about 90 mA running Popcorn, but 10-20mA when it wasn't working hard.)
    •  
  • I also mention of adding capacitors, to either or both of, the input or output of voltage regulators. Such components are often referred to 'decoupling' capacitors. Their job is to act as a small power reservoir for transient current demands. Although these current demands can be quite high, they are so short they do not cause immediate component failures ... although premature failures over months and years is feasible. They help with the general stability of the system, but omitting them is very unlikely to result in component failures. The exact values are not usually critical, with 10 uF to 100 uF being common choices for an electrolytic. Electrolytic capacitors are efficient in providing a high capacitance, but their function is degraded at very high frequencies. Hence, it is common to parallel a ceramic capacitor in the value range 10nF to 100nF, to act as 'high frequency helper'. Wirelength also plays an important part ... if the capacitance is 'too far' from the device causing the transient current demand, the inductance of the intervening wire negates the effect of the capacitor. 'Too far' is difficult to quantify .. it depends upon the speed of the circuit ... in some cases it can be centimetres ... in others just a millimetre or two. Hence, it is common to place many capacitors in parallel across the power lines.  Whilst there are exceptions, the usual rule is too many is not a problem, too few might be a problem.

I hope this helps a little. Power supplies are the 'utility' part of the project, but they are critical and more complex than they might appear.

Best wishes, Dave


   
Inst-Tech reacted
ReplyQuote
Will
 Will
(@will)
Member
Joined: 3 years ago
Posts: 2504
 

@davee 

Thanks for the overview, we agree with you in that we've come to the general consensus that the LMS1117 module is too wimpy for this job.

Anything seems possible when you don't know what you're talking about.


   
Inst-Tech reacted
ReplyQuote
(@davee)
Member
Joined: 3 years ago
Posts: 1609
 

Hi @will ,

re: LMS1117 module is too wimpy for this job.

As a general purpose solution, then yes, it is at best marginal, for a 12V to 5V job. Bill was sensible in picking a little buck supply.

However, looking at the datasheet alone (I haven't tried it), if the Pico was 'not busy' and there are not significant current hogs (sensors, motors, etc) attached, then it looks as if it should work. (I am thinking of something like blinky as a total task.)

If the AM1117 are frying instantly, there is another problem lurking, and it would be wise to find that before committing to more expense, etc.

Best wishes, Dave


   
Inst-Tech reacted
ReplyQuote
Will
 Will
(@will)
Member
Joined: 3 years ago
Posts: 2504
 
Posted by: @davee

However, looking at the datasheet alone (I haven't tried it), if the Pico was 'not busy' and there are not significant current hogs (sensors, motors, etc) attached, then it looks as if it should work. (I am thinking of something like blinky as a total task.)

If it were only a blinky the whole thing could be replaced with a flashing LED and a suitable resistor 🙂

If the AM1117 are frying instantly, there is another problem lurking, and it would be wise to find that before committing to more expense, etc.

No pyrotechnics here. The OP described it as ...

"Here's what seems to happen a lot (but not all) of the time. I have the complete RPi Pico Pack assembled as per the schematic. I plug in the USB, and all is well. I unplug the USB (which isn't really necessary due to the presence of D1,) and then plug in the external 12V supply. Some of the time, all is well, but some of the time, the AMS1117-5 Module fails. The only reason that I have thought of that seems logical is that the RPi Pico draws more current at startup than the AMS1117-5 module can supply."

Anything seems possible when you don't know what you're talking about.


   
Inst-Tech reacted
ReplyQuote
(@davee)
Member
Joined: 3 years ago
Posts: 1609
 

Hi @will 

some of the time, the AMS1117-5 Module fails

   I don't know what 'fails' means in this context ... sometimes things quietly demise of (premature) old age, but power things usually get hot .... so I am assuming they are frying .. perhaps not ... they may just be a batch of junk ... but the ones I have seen, tend to work...

Hence, I wondering if they are failing of their own accord ... or are they being pushed?

............

My suggestion of blinky or similar was not as final project ... just as a test to see if it provokes a failure! When a board is hooked up to other stuff, the other stuff could be causing a problem.

-----------

I haven't yet fired up a Pico, but I don't believe it is likely to have any major current demands ... even a small peak should not cause the AM1117 to fail.... at worst the Pico power would dip and it would reboot or similar.

For my first suspect, I would look at the output from the 12V supply. It could be innocent, but with wiring and hardware, I treat everything as guilty until proved innocent.

Then as second suspect, anything attached to the Pico.....

Best wishes, Dave


   
Inst-Tech reacted
ReplyQuote
Inst-Tech
(@inst-tech)
Member
Joined: 2 years ago
Posts: 554
 

@davee Indeed, well said @davee!

to expand on your summary of why it is important to revue the specifications of a device,

I put forth a scenario of using a 9v supply to a AMS1117-5, ( 5v output linear regulator)

to calculate the heat dissipation of the device at ambient temperature of 25°c.

Even those that are math challenged will hopefully find this interesting,

regards,

LouisR

LouisR


   
DaveE reacted
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 3 years ago
Posts: 6662
 

@tracecom How goes the experiment with the MINI-360? Also, does your product have a name yet? Is it on Kickstarter?

First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, and 360, fairly knowledge in PC plus numerous MPU's and MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
Sure you can learn to be a programmer, it will take the same amount of time for me to learn to be a Doctor.


   
ReplyQuote
(@tracecom)
Member
Joined: 2 years ago
Posts: 28
 

@zander Mini360s don't arrive until sometime tomorrow.


   
Ron reacted
ReplyQuote
(@tracecom)
Member
Joined: 2 years ago
Posts: 28
 

I made a couple of video clips, but they are too big to post here, so attached are a couple of photos. One shows the AMS1117-5 successfully powering the RPi Pico Pack and the other shows the Mini360 fly-wired in and also working well. After the last few days of reading all your comments and performing some testing, I am satisfied that the Mini360 is the better choice. I have redesigned the PCB to accommodate the Mini360 and make some other improvements, and will post again here in a week or so when I get the new boards and get one assembled. Thanks to all who helped.

RPi Pico Pack with AMS1117 5
RPi Pico Pack with Mini360

   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 3 years ago
Posts: 6662
 

@tracecom Are you intentionally keeping the product a secret from us?

First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, and 360, fairly knowledge in PC plus numerous MPU's and MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
Sure you can learn to be a programmer, it will take the same amount of time for me to learn to be a Doctor.


   
ReplyQuote
(@tracecom)
Member
Joined: 2 years ago
Posts: 28
 

@zander I posted pictures of it, stated that it was based on the project that Bill made the video and wrote the article about, posted my schematic diagram, described the problem I was having, and answered lots of questions that you asked. If I was trying to keep it a secret, I certainly didn't do a good job of it. I told you exactly how much the PCBs cost, told you my YouTube channel. How much more do you want to know? and why?


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 3 years ago
Posts: 6662
 

@tracecom The reason is I built my own solar system for my RV so when you mentioned RV I was curious, especially the unusual 12.8V. This thread has gone on so long I have forget where it started. If you told us your YT channel, I have forgotten it, can you supply a link?

First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, and 360, fairly knowledge in PC plus numerous MPU's and MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
Sure you can learn to be a programmer, it will take the same amount of time for me to learn to be a Doctor.


   
ReplyQuote
(@tracecom)
Member
Joined: 2 years ago
Posts: 28
 

Actually, I feel like you are intentionally trying to annoy me, and it's working. This thread is not about RV's or solar systems, and I don't want to debate batteries with you. And that certainly is no reason for you to accuse me of keeping secrets. You seem to want to demonstrate that you know more about everything than I do, and you may be right. But I do notice that you seem to go off subject quite often. So my suggestion is that if you want to know more about my project, you reread this thread and then Google whatever else you want to know. I wish you well, but I am finished explaining. Good bye.


   
ReplyQuote
Ron
 Ron
(@zander)
Father of a miniature Wookie
Joined: 3 years ago
Posts: 6662
 

@tracecom I am sorry you feel that way. The reason that RV's or solar systems came into the discussion is because you said https://forum.dronebotworkshop.com/postid/32470/

First computer 1959. Retired from my own computer company 2004.
Hardware - Expert in 1401, and 360, fairly knowledge in PC plus numerous MPU's and MCU's
Major Languages - Machine language, 360 Macro Assembler, Intel Assembler, PL/I and PL1, Pascal, Basic, C plus numerous job control and scripting languages.
Sure you can learn to be a programmer, it will take the same amount of time for me to learn to be a Doctor.


   
ReplyQuote
Page 5 / 6