Announcement

Collapse
No announcement yet.

Dyson Cyclone v10 - Brushes not working

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Dyson Cyclone v10 - Brushes not working

    Hey everyone,
    I have this portable vacuum cleaner that works just fine, except for the attached brushes that are completely dead. The vacuum motor has 3 speeds, but the brushes always worked at the same speed regardless. After a quick check I realized that the brushes are not getting a way to ground when the cleaner is on.

    I accessed the motherboard and started probing around. Unfortunately, I didn't fully remove the motherboard because my iron is not powerful enough to melt the solder connection to the motor (vacuum) in the center of the board. The board was just soaking all the heat and I felt like I could possibly damage things even more if I persisted, but there are components on the other side as well.

    The circuitry for the brushes seems simple enough (see attachment), Q16 seems like the brushes' way to ground, maybe an N-channel mosfet or an NPN transistor. I googled the numbers with little results. While on, it's not receiving any signal on its base/collector (0v). I couldn't troubleshoot any further since the trace goes to the other side of the board, but it measures about 60k (rising up to 90k) to B+.

    There might be more to the mystery, but with no schematic and no access to the other side there's no way to tell.

    The battery measures around 27v and has its own separate protection and charging circuit. I also tested the brushes with an external constant voltage power supply, they seem to work just fine at around 8-9v, I could imagine they'd go crazy with 27v, that's why I tend to believe that Q16 could be a transistor.

    Now here for the questions:

    1. Would it be safe to soak the board with heat in an effort to desolder the motor connections and access the other side of the board for further troubleshooting?

    2. If I could identify Q16, would it be crazy to just cut the trace to the original signal and feed it an external one with the appropriate resistor configuration?

    Thanks for any help and suggestions, if you need any more info or pictures just ask.






    Attached Files

    #2
    Re: Dyson Cyclone v10 - Brushes not working

    Just a little update on this repair in case anyone will have the same problem in the future.
    I managed to remove the board but gave up on tracing the problem, the whole board is covered with an insulating material and without a datasheet I simply gave up.

    So I decided to bypass the circuit that controls the brushes. After some trial and errors and a little bit of learning I came to the conclusion that the brushes are controlled with PWM signal. The bushes are DC 16.2v, 50W.

    I understand very little about the right frequency for motors, so I went by ear and found that what resembles the original sound the most at 27v is 2KHz at 20-30% duty cycle. I created a small board to bypass the circuitry with an AMS1117, Attiny13 and a K355 N-mosfet sourced from an old laptop motherboard and connected everything to B+, B- and Brushes- (diode is still connected and working from the motherboard).

    Everything fits and works, however, when releasing the button from the vacuum cleaner, the battery pack takes 4-5 second to disconnect. The vacuum itself stops immediately after I release the button, but since my board is powered directly from 'battery IN', it takes about 4-5 seconds for the brushes to stop moving.

    I'm going to disassemble everything and hook an opto directly to the motor (judging from the signal looks like it's an AC motor, with no common ground with B-), and use the output to control my PWM signal.

    I'll post updates with the full repair, hoping I won't destroy anything in the meantime. For anyone with the same problem, you could also get a step down DC-DC transformer ( 5 amp just to stay safe) and hook the input directly to B- and B+ and the output to the brushes (don't forget to disconnect the brushes input from the motherboard and add a diode!!!!). Make a couple holes and fix it outside the vacuum cleaner in a good spot, and don't forget to insulate it. The brushes will still take 4-5 seconds to stop after you release the button, but at least it will be a quick and easy fix.

    Attached Files

    Comment

    Working...
    X