Magic Smoke


 
Well, I got the MOSFET out. The continuity checks out on both traces (see photos below). And, the fan spins when connecting it directly to the 10v power supply.

g7dyyQVl.jpg


1X3gbbZl.jpg


Did I check everything you asked me to check?
 
Yeah nice job taking that part out! I think that part's gotta be it. It is the only one I see any stress on if I short the output. The traces look good so hopefully just the fet holds all the smoke you need.

Did you want to get a different case or do you just want me to send the mosfet? You can just email me and let me know what you want.
 
I received additional parts from Bryan (special thanks). Ralph, I got the extra BS170 for Q1 or Q2 and the Inductor. I intend to install them if/when it is necessary. But, for the sake of troubleshooting and identifying whatever issues I may have, I have only installed the new FQU MOSFET in Q3, so far. When the problems are corrected, I believe my issues will be beneficial to others. So, I want to be able to identify the source of my problem, if possible.

So, after installing the new FQU MOSFET, my current issue is that every time I connect the red lead of the servo to the green/white 5v lead of the cable, the HM buzzer beeps, LEDs flash and everything kinda goes crazy. I think the Pi is rebooting, too. With the fan connected, but without the servo being connected, it's all good. I can put the fan in manual mode and ramp the fan speed up and down. But, if I try to connect the servo to the cable, everything goes nuts. I tried a spare servo that I have and get the same results. The servo motors do start moving when I touch the wires.

I checked voltage on the cable and I am getting 4.97v on the green/white lead and .14v on the green. The servo and fan share the blue ground.

This next part should probably be a new thread because it has nothing to do with the servo problems. Bryan, just tell me if you want me to start a new thread with the following. Anyway, before I could start testing my new MOSFET, I ran into wifi issues. I'm wondering if the wifi problems aren't an issue with the latest image because the only way I was able to work around it was to install an older image. Here's the details on that:

Heatermeter Steps (wifi issues - worked around):


  • Right after getting the new MOSFET installed and rebooting, I cycled through the menu using the buttons to get to manual fan mode
  • Somehow it started resetting the config, I don't know if I pressed the "Next" button too quickly or what, all I know is it said, "Resetting config..."
  • After it rebooted, the thermocouple reading was off because it had been reset to thermister
  • While resetting it back to thermocouple, the web gui informed me that no root password was set
  • I tried resetting the passwd via the GUI, but it wouldn't let me
  • I tried via SSH. It said file system was read only
  • So, I updated the linkmeter firmware
  • Still couldn't reset via webui
  • Decided to go to wiki and download the most recent OS image
  • Installed it and put my Wifi settings in the config.txt, just like I did before, but it kept ignoring my settings.
  • I tried re-imaging 2 or 3 times, once with "zcat", once with "dd" and rebooted into Windows used Win32 Disk Imager. I fought with it for a long time.
  • Finally I decided to put my older image on, put wifi settings in config.txt and Wifi worked first boot
 
Sounds like you're getting brownouts on the power supply, or there is a short. Either way, the power supply is cutting out. Have you gotten a power supply yet, either a 12V 1A name brand or 12V 2A chinese noname?
 
Hmmm. It appears my power supply was an issue. I was using the 9v 1A that, iirc, was putting out 10'ish volts. The one that we mentioned the fan wouldn't function correctly. But, I found another one to try. It's a 12v 1A that's putting out just a little over 12v. I just tried it and I didn't get the "ape sh!t craziness", when I briefly touched the wires. The servo made a noise and moved and nothing beeped. I'm a little nervous about leaving it connected now, lol. I guess it's okay. Do you advise that I just go ahead and wire it up?
 
You'd have saved yourself a lot of effort and confusion if if you'd just started with a power supply that meets the stated requirements! Sounds like you're where you need to be now though, although it depends on the quality of that adapter.
 
Ohms law states Power = Voltage X Current. If you want to deliver the same power with 9V that you deliver with 12 you can see easily that the current must go up. If you have a PS that is under voltage and under rated for current you will get brown outs. Though Bryan's recent rundown of power supplies seems to indicate 12V 1A should hold solid I had brown out issues with a 12V 1A cheapo amazon supply as have many others in the forum. You are at 9V to start, asking for trouble IMHO, use a proper supply to eliminate headaches.

Hard to tell what you have right or wrong with your servo. What are your Servo Pulse Duration settings, and what is the Servo Fully open setting set to? First thing you want to do is set those items to a reasonable range, 700-2000 should be almost full rotation for the servo and I would set it Fully Open at 100%, this will give you a lot of motion on the servo as you run the HM from 0-100%.

As for the servo circuit on the HM, it is very simple. The servo just draws 5V from the HM to power it and then the input signal wire comes from the ATMega through one BS170 and that's it. In fact, the servo wire used to connect directly to the ATMega without any additional circuitry, the BS170 is just a little servo booster to help keep the servo steady over longer cable runs. I wouldn't suspect any issues in this circuit as long as the soldering is solid.
 
Thanks guys. The power adapter that I was using is the one I was using with the HM 4.1. That's why I selected it. I thought it was "proven".

I am so appreciative your generosity. And, I just can't thank you guys enough. I apologize for any inconvenience that it caused you. As for me, it's a learning experience. And, maybe others will learn from my experience, as well.

I checked the voltage of my current power supply. It's putting out around 12.3v. I have never checked amperage before. But, if I did it right, it looks like it's measuring about 1.2A (or, more specifically, 1.5-1.8A). Does this sound like a "proper" power supply? I don't mind replacing it with a more reliable power supply if you think it's warranted. I didn't realize I was "cheaping out" with the original (21v) power supply and don't want to make the same mistake.

These are my current settings. I lowered the 1000-2000 Servo Pulse Duration to 700-2000 per your suggestion. The PID should be stock since I've re-imaged. Right now, I'm just trying to get it all wired up and haven't thought about them.

REv7dPKl.png
 
I was referring to what the PS is rated at, look at what is printed on the case... I think you said yours was 9V, and you were having brown out type symptoms? If your ps is rated at 12VDC / 1.25Amp that should be fine. If you are saying you are measuring 12.3V that is perfect, but if you are measuring 1.2A of current that is a lot of draw, more than a properly functioning HM should be pulling unless it was spinning the hell out of a large blower....
Your settings look fine, I only mentioned the SPD settings because I have seen people set them at 1000-1000 and wonder why their servo isn't moving. Is your servo moving now or are is your HM still going spastic?
 
I think the newest power supply (replacing the brown out 9v) is rated 12v and 1A. I measured the voltage and amperage without it connected. I have been tied up all day (still tied up) and won't get a chance to reassemble until tomorrow. I will post back.
 
I just finished reassembling everything. I'm getting motion out of the servo now. And, the fan is moving. That's as far as I've gotten. But, I wanted to post and tell you guys about my progress and thank you for your help. I am so grateful for all of the help. Thanks so much.

I also want to recap for anyone who might be following the thread what I believe happened:

First, after the initial assembly, I couldn't get motion out of the servo. Then, while troubleshooting, I *think* I might have shorted the fan or servo lead with my multimeter probe, burning up the FQU. I'm not 100% how that happened actually. It could have just been my power supply took its toll on it. It appears that the reason that my servo wasn't working was because my power supply was putting out close to 21v. It was rated at 16v, which I thought was safe and was the same power supply I was using on my HM v4.1. After replacing the FQU and finding a better power supply, I was able to wire everything up. I believe it is all working at this point.
 
Good to hear it is working. If you didn't replace the inductor you should test the blower in in Voltage mode to make sure it is ramping properly. Set the blower Max to 100%, in manual fan mode step up from 0 to 100% 5 degrees at a time and make sure it spins from a trickle to full blast. If the inductor is toast it will probably jump from off to spinning about 80% in the first 10%. If you suspect something isnt right with fan speed switch back and forth between voltage and pulse mode and see if the fan is spinning roughly the same speed.
 
Thanks for the tip, Ralph. I will give that a try as soon as I can and post back. It will probably be tomorrow evening after work
 
Welp, you called it again. Looks like I'm going to need to replace it, as well. You're a genius, Ralph :-). Thankfully, Bryan included one in my last order. I never replaced the other BS170 either. Can it be assumed to be good? You know, I wanted to only replace things one at a time so that I could A) know what was broke and B) not risk doing additional damage. I just wanted to make that point so you'd know I wasn't ignoring your suggestion to replace the other parts.
 
If the blower is going on and off with the HM commands then the BS170 should be OK. The inductor is in the feedback circuit which is used by the HM in voltage mode to regulate the blower speed, so it wont effect things when in Pulse mode (where the HM doesnt need the feedback to regulate the blower speed)
The reason I was so adamant about the inductor is because its malfunction is easy to miss (because the blower will still work but the speed regulation will be way off). It took me a while to diagnose and fix the one I had with the blown inductor, if there is enough current flowing to blow the FQU that current has to pass through the inductor too, so it is likely toast too.
At any rate, easy to fix and you should be good to go when that is done..
 
You must log in or register to reply here.

 

Back
Top