Bowflex nautilus firmware woes

[CuriousJohn]

Ok cool, I think i get the picture now it sounds like a short as 4 ohm across C79 is way to low it should be 190 ohm or so, so something fed by U18 is suffering a short out to ground :-). unlikely to be C79 itself, easy to pop off with iron or better still hot air, check U18 if U17 is feeding it 3.25v maybe 3v not less really as its op should be 2.85v. I would defo try the IPA again to try localize to the sc, do it in small areas try on U17 for sure it should be hot it feeds U18 which should also be hot and maybe even caput.

For my one it turned out to be U14 the PIC32 MCU short to ground having been cooked by a laptop psu but before i removed U14 i did the IPA test on it and watched it steam off at power on and removed U18 to see if anything other than U14 was stressing and found the U17 Buk that feeds it still hot and tired even with U18 off the pcb so replaced that component to mostly using hot air gun carefully its possible to do with an iron if you swish the tip with loads of solder on both sides prying it up as you go but U17 Buk really needs hot air as it has a pad underneath as well between pcb and itself.

[CuriousJohn]

Nothing to be sorry about I am not really educated in board repair beyond watching Lois Rossman! Im just a curious idiot with a volt meter. I do appreciate you humoring my efforts.

The photo I attached was merely to show that some tachometer LEDs do work albeit are very dim - the panel stays that way with no changes - had it on for 30+min to see maybe it will do something. When shop lights are on, you cant tell there is any power - it certainly does not light up like in your photo... maybe 1/16th of that. No life from green LEDs at all - I did have all ribbon cables hooked up tight.

I have tested backlight with 3V coin battery and both sides light up fine - tested one at a time. One would think having some voltage to there(2.4 currently) would at least allow the backlight to glow... which it isnt without using the 3V battery.

Measured resistance between ground and the cap indicated and get 1ohm. Top side gets 4ohms.

Heres video of it booting takes a second to see anything... Its certainly not as bright regardless of lighting. Looking at the boot requence mine is stuck where inner tachometer LEDs are faintly on. Left it up and running and screen reads same garble. Nothing on 7 segment LED either.

https://youtube.com/shorts/NAdPOw99I...nhZHkRh9sC72K7

Left it on for about 10 min and screen had not changed nor I get any button response. Nothing legible there. It almost as if its not getting the power it needs to boot fully...

Sometimes I think im an optimist but often hit a wall that confirms realism is my true calling. I hope you dont give up with it but also dont waste weeks futilely fighting something that cannot be defeated in this lifetime lol. At the very least you could do the manual resistance conversion so the max trainer can be used again and maybe add a generic exercise bike computer for the lolz.

IMO your console screen is not as broken as it might seem, i think it boots because of the behavior the LCD appears to power on self tests ok (this is just the LCD driver chip testing so that bit is fine) but cannot light its tacho leds and lcd backlight or display on its lcd screen properly once booted because the short circuit on the o/p out of U18 has caused problems getting the serial data into the lcd screen chip as logic level to low sort of thing or maybe even the low volts disrupting that LCD driver chip enough to cause it not to run after its self test passes? The leds including the lcd backlight dont try to light at all because the PIC32 controls the cathode sink so they never get told to switch on.

I suspect your BLE module is still working and thats important because its the only part thats irreplaceable as the firmware is not available because it cannot be extracted being locked at write time for security i guess so nobody can clone it and start producing new machines. So if it is just the PIC32 that is short circuit caput I managed to extract the firmware from that PIC32 MCU and write it back to a new chip but to replace it will need medium skill with hot air gun and the associated cost of getting a hot air station whilst not massive is enough to be put off unless you know somebody locally who could oblige. To save you the cost of getting a PIC32 programmer I could supply you with the PIC32 chip already written ready to go on the pcb but again shipping costs to consider from UK and a pic32 programmer would probably be equal to the cost of shipping the chip top you anyway. The firmware file link is earlier in this thread somewhere, i uploaded it to archive.org.

All that said U18 voltage regulator supplies many of the chips on the pcb, most are listed on the power distribution diagram on post 15 in this thread so it could be any one of those components that have a short or even any capacitor associated with any of those chips power in pin really but imo the PIC32 is most likely to get fried first in any over voltage situation as it has 4 pins to U18 voltage supply and another 4 pins to ground which means it has the path of least resistance to ground where any excess current would break through first imo but it might be one of the other chips especially any of the 5 shift registers U7,U12,U10,U9,U8 or the SPI flash chip U13.

If you dont intend to crack on with it I might be interested it taking it off your hands to find out, you know where to find me so have fun!

[stj]

Nothing to be sorry about I am not really educated in board repair beyond watching Lois Rossman! Im just a curious idiot with a volt meter.

good places to learn stuff.
https://www.youtube.com/c/NorthridgeFix/videos
https://www.youtube.com/user/bigclivedotcom/videos
https://www.youtube.com/channel/UCQa...B_Kd54g/videos

[CuriousJohn]

good places to learn stuff.
https://www.youtube.com/c/NorthridgeFix/videos
https://www.youtube.com/user/bigclivedotcom/videos
https://www.youtube.com/channel/UCQa...B_Kd54g/videos

cheers, Ill have a look as well lol also being a curious idiot haha.

[CuriousJohn]

Bowflex Max M7 console logic stuff diagram and backlight/led stuff.

LCD backlight is turned on by switching its cathode sink switches Q2+Q3 ip high for on and low for off but this comes from the ip side of a buffer chip U4 and i am yet to find where this actually feeds from I suspect its the PIC32 or something here is not understood. Can a none inverting buffer chip also behave as a sink so its reversed and its o/p state informs its ip state by holding it high? Meh

Please disregard the post above it is incorrect and rambling it was late and i was tired. 😑

I have just spend a whole 5 minutes this afternoon figuring out that pin 28 of U14 the PIC32 MCU, switches the LCD backlight on and off by driving its cathode sink transistors and there are two sets as two strings in the led backlight by looks of it and that signal high or low from P28 also goes into input 2 of U4, a none inverting buffer thus turning the tacho leds on and off and the 7 segment display as that is a common cathode part so all 4 cathodes are controlled by shift register as well as are the cathodes of the tacho leds in a similar way. All 5 shift registers are daisy chained together and the OE enable signal from p28 via the NI buffer U4 is common to all of them. Sorted. :-)

Updated logic diagram below its a bit busy sorry its more my notes than owt else.

[CuriousJohn]

I have tested backlight with 3V coin battery and both sides light up fine - tested one at a time. One would think having some voltage to there(2.4 currently) would at least allow the backlight to glow... which it isnt without using the 3V battery.

2.4v is a bit low should be 3v or over, this is coming from from the output of our U17 BUK now stressed because of the fault and U17 also supplies 3v to U18 the voltage regulator also under pressure from the sc that is suspected on U14 PIC32. U14 also turns the backlights and leds on by Pin 28 benig high or low as needed by its program.

The cathode of the backlight leds are switched to ground by two sets of two transistors. The anode supply you measured as 2.4v is always on but should be 3v. The cathode sink that switches them lights on and off by controlling the path to ground for each led string in the backlight and to repeat the signal for that is pin28 of U14 which also feeds into input 2 of a none inverting buffer U4 to control the cathodes of all tacho leds and 7 segment display by the shift registers.

[CuriousJohn]

Logic diagram update with corrected info.

[CuriousJohn]

-------------------------M7 Hardware Mainboard---------------------------------

U1 N2211 - NCS2211D Class A/B audio amplifier.

U2 WM8524G - 24-bit 192kHz Stereo DAC

U3 C345 - SOT 23-6 SN74LVC2G34DBV CMOS logic IC Dual noniverting buffers
U4 C345 - SOT 23-6 SN74LVC2G34DBV CMOS logic IC Dual noniverting buffers

U5 Bluetooth module - BLE112-A TI CC2540F128

U6 PT6578LQ - LQFP 100 pin - LCD driver : https://www.princeton.com.tw/en-us/P...ment-Driver-IC

U7 74HC595D - Texas instrument 8 bit shift register SOIC 16 narrow (pdso g16)
U8 74HC595D
U9 74HC595D
U10 74HC595D
U12 74HC595D

U11 GF2481GS quad 7 segment LED - common cathode 6,7,10,11
Q27 72k sot23 - ELM98030BC 3.0V±2%, 200mA Linear Voltage Regulator IC
Q28
Q29
Q30

Q39 K84 sot23-3 - BSS84 P−channel enhancement−mode fet

Q7,9,11 HF smd sot23-3 -

Q2,Q4 s8050 c-j35

U13 25L8006E - 8Mb NOR spi flash (4Mb x 2)

U14 PIC32MX470F512H - https://www.mouser.co.uk/ProductDeta...X470F512H120PT

U15 e78e 725m87 - AS78L05RTR-E1 100mA POSITIVE VOLTAGE REGULATOR

U16 GAF smd 5 - RP170N381B Linear voltage regulator IC

U17 cxw8509 - Synchronous Buck Regulator 4v - 25v :-)

U18 L1RC smd - linear regulator ADP3309ARTZ-2.85 OR ME6221CM5G

U19 WT7901 - Motor driver

Q33 C2655 - 2SC2655 NPN Transistor

U20 GH23H smd 4 - AZ1117EH-ADJTRG1 1.0A Low Dropout Linear Regulator

U21 L1RC smd - linear regulator ADP3309ARTZ-2.85 OR ME6221CM5G

U22 C04F - SOT23-6 SN74LVC2G04DBV - dual inverter
U23 C04F - SOT23-6 SN74LVC2G04DBV - dual inverter

U24 p8563 05 GI - PCF8563TS/5 Real time clock module
C98 kamcap super capacitor 5.5v 0.33f

U25 SN74LVC125A - quad gate driver SOIC 14 (14 pin)

[CuriousJohn]

I have tested backlight with 3V coin battery and both sides light up fine - tested one at a time. One would think having some voltage to there(2.4 currently) would at least allow the backlight to glow... which it isnt without using the 3V battery.

It can be daunting looking at a complex diagram so its best to take it in small bites so i offer this for understanding of the backlight.

The backlight has two "strings of leds" so at least two different leds are in it, focusing on one side only but both sides work identical;

The anode supply at J12 is always on 3v supplied from U17 BUK, U18 powers U14 (2.85v) and U14 pin 28 turns transistors Q5+Q4 on and this allows the backlight leds to illuminate as current can flow through the leds to ground via Q4 + R48.

[Zhinox]

Hello everyone
I know absolutely nothing about electrical engineering. But briefly on this topic. I have a Bowflex M8 with the same problem. The console doesn't start when I plug in the power cable. The L1RC on the circuit board then gets extremely hot. However, if I pull out the plug at the bottom of the servo (blue box), the console works and the L1RC gets cold. I can then plug in the servo and the device works. If I unplug the power cable and plug it back in again, the problem starts all over again. You may have the same problem and this could help solve it.

[CuriousJohn]

Hi Zhinox. The M8 console may be slightly different to the M7 internally, i know M8 can use the JRNY app so has different technical abilitys to the M7 I would be very interested to see a decent quality photo of the inside of it if you have the back off still.

As for your fault you say L1RC gets hot, is that labelled as U18?

This fault does sound a bit funky which leads me to wonder if the supercap is failing, after a day or two maybe a week being powered off, does it ask for the date and want to setup a user profile like its the first switch on? If its behaving like that almost certainly a failing supercap which is a fancy battery that provides power to store the date info in the Real time clock it can cause power surge at start if failing, but to test the supercap a voltmeter or multimeter would be handy to measure the voltage across it when the console is both off and on. The supercap should charge to around 4.3v when on and keep that value when off whilst slowly dropping a bit, see the videos below for more on supercap.

Bowflex M7 console Bad supercap
https://youtube.com/shorts/dvxnN5aw4vM?feature=share

Bowflex M7 console supercap removed
https://youtube.com/shorts/hXes2dojA5w?feature=share

Bowflex M7 console New Supercap
https://youtube.com/shorts/YV182CAHDKY?feature=share


Another cause then and more likely for you given the scenerio described is the potentionmeter or "pot" in the servo motor, it is the posh name for a variable resistor, they are known to drift out of tolerance causing issues.

The symptom is erratic variation in resistance brake position with relation to the resistance setting on the console. The problem is caused as the "pot" which is a 5K ohm (5000 ohms) variable resistor, due to wear or failure it drops in value to below 4K ohm even lower, the last one i repaired the resistance measured 3K ohm with the ohm meter on my multimeter which is way out and caused the console to drive the servo motor hard as it couldnt find the voltage level it wanted to find for its position. These pots are supposed to remain within 10% of there stated value "tolerance" but are falling way out of that causing issues. Would it cause your fault? Maybe if the servo motor was being driven hard causing lots of current to flow which might explain your hot voltage regulator U18? This can also wreck the gears inside so defo worthwhile to measure the resistance of the pot, it should be 5K ohm or 10% either side so 4.5K ohm to 5.5K ohm, anything outside this range means its fallen out of tolerance and will likely get worse.

Photo of Servo internals for info, you shouldnt need to open it up to measure the resistance but will need to open it if new pot is needed. The pot is down the bottom with 3 wires going into it. The resistance between the two outer wires should be 5K Ohm, measure the resistance of the pot with the resistance setting of a multimeter or ohm meter if you have that.





The pot on this one is not covered in hot glue its easy to see the terminals on its back. The resistance between the red wire and black wire should be 5K Ohm or within 10%.





Dont rule out having both these faults or even none of them lol but a multimeter is needed for quick simple measurment of both DC voltage and resistance in order to determine the status of both the supercap and servo pot.

[Zhinox]

Hello John
It's funny that you write something about the super capacitor. It was defective when I received the device. See photo above left. I replaced it with a new one. It could also be that the new super capacitor is causing problems.
The L1RC is U17. See photo red arrow.

The super capacitor has a little over 4.5 volts when the device is switched on. And it is additionally charged. When you unplug the device, the super capacitor loses voltage. The servo was unplugged during the test.

I measured the servo without power, so all cables are disconnected. If I measure both outer pins, I get 5.18 kOhm

[stj]

super capacitors - also known by other names are a failed design.
they only survive a limited number of cycles and the cheaper ones in aluminium cans like to leak and wreck pcb traces.
the better ones from ELNA and NEC are steel-cased but still have a limited lifespan

if it's been replaced then check near it for damage from the old one - specially throughhole traces.
it's gets into the holes and eats the sleeve away

[CuriousJohn]

The Servo seems ok then, thats good news you didnt mention it going out of resistance setting when set so assumes it doesnt. It would be an idea to check the resistance from the middle pin to each outer pin they should both be different readings unless you managed to put the pot dead centre lol but both readings will add up to the total you measured of 5.18K or thereabouts, just to confirm the pot is fully working as it should for completeness if you care to lol.

Did you say the supercap you put in does not hold 4.3v after powering off but drops rapidly after removing power? if so the super cap may be bad especially If fault started after putting new super cap in, probably a good tell its a problem. Also as mentioned by "STJ" its possible some damage may have occured especially removing old one as i do recall they are a pain to get out the solder is tricky to remove without a desolder iron.

The super cap just provides power to RTC to store date for JRNY app? If you dont use JRNY and can tolerate the startup message every time then do you need it?


I have updated the M7 console diagram as needed a few errors correcting see below. It has RTC + Supercap circuit in centre its same as on M8 just in different location thanks for posting photo. I would check D9 resistance to make sure its not gone short (diode testing) and R170 (R150 on M7) although be mindful doing this with supercap installed and charged might produce strange resistance readings as mayeb confuse meter best to check when removed supercap or could snip it out.

So on M8 U17 looks like the MCU main regulator, its in the same location as U18 on M7 console board doing same job i think. I thought this regulator fed the RTC and supercap but that was wrong the RTC and supercap is charged from U17 on the M7, lookslike U16 on M8, easy to check continuity between output of U16 to D9 it should be near 0 ohm as just a copper trace connecting them. Or with 9v being fed onto the board at J17 pin 8 (9v) and pin 9 (gnd) the output of U16 should be 5v and that goes into the anode of D9 among other places lol.

Any chance you could do a better photo of the U12 & U10 area as keen to know what those chips are, not sure if U10 is a realtec or raspberry pi something but i doubt lol. that board defo has a different Blutooth module which might explain the JRNY compatability thing as M7 is not JRNY compatible.


Updated power Distribution of M7 console.




------------------------M7 Hardware Mainboard---------------------------------

D9 SOT23 "JV" single diode package

U1 N2211 - NCS2211D Class A/B audio amplifier.

U2 WM8524G - 24-bit 192kHz Stereo DAC

U3 C345 - SOT 23-6 SN74LVC2G34DBV CMOS logic IC Dual noniverting buffers
U4 C345 - SOT 23-6 SN74LVC2G34DBV CMOS logic IC Dual noniverting buffers

U5 Bluetooth module - BLE112-A TI CC2540F128

U6 PT6578LQ - LQFP 100 pin - LCD driver : https://www.princeton.com.tw/en-us/P...ment-Driver-IC

U7 74HC595D - Texas instrument 8 bit shift register SOIC 16 narrow (pdso g16)
U8 74HC595D
U9 74HC595D
U10 74HC595D
U12 74HC595D

U11 GF2481GS quad 7 segment LED - common cathode 6,7,10,11
Q27 72k sot23 - ELM98030BC 3.0V±2%, 200mA Linear Voltage Regulator IC
Q28
Q29
Q30

Q39 K84 sot23-3 - BSS84 P−channel enhancement−mode fet

Q7,9,11 HF smd sot23-3 -

Q2,Q4 s8050 c-j35

U13 25L8006E - 8Mb NOR spi flash (4Mb x 2)

U14 PIC32MX470F512H - https://www.mouser.co.uk/ProductDeta...X470F512H120PT

U15 e78e 725m87 - AS78L05RTR-E1 100mA POSITIVE VOLTAGE REGULATOR

U16 GAF smd 5 - RP170N381B Linear voltage regulator IC

U17 cxw8509 - Synchronous Buck Regulator 4v - 25v :-)

U18 L1RC smd - linear regulator ADP3309ARTZ-2.85 OR ME6221CM5G

U19 WT7901 - Motor driver

Q33 C2655 - 2SC2655 NPN Transistor

U20 GH23H smd 4 - AZ1117EH-ADJTRG1 1.0A Low Dropout Linear Regulator

U21 L1RC smd - linear regulator ADP3309ARTZ-2.85 OR ME6221CM5G

U22 C04F - SOT23-6 SN74LVC2G04DBV - dual inverter
U23 C04F - SOT23-6 SN74LVC2G04DBV - dual inverter

U24 p8563 05 GI - PCF8563TS/5 Real time clock module
C98 kamcap super capacitor 5.5v 0.33f

U25 SN74LVC125A - quad gate driver SOIC 14 (14 pin)

[CuriousJohn]

super capacitors - also known by other names are a failed design.
they only survive a limited number of cycles and the cheaper ones in aluminium cans like to leak and wreck pcb traces.
the better ones from ELNA and NEC are steel-cased but still have a limited lifespan

if it's been replaced then check near it for damage from the old one - specially throughhole traces.
it's gets into the holes and eats the sleeve away

these boards where my first real dealing with super caps so thanks for this.

For the record I bought replacements of ebay, they are named "Powerstor" so far seem ok : 0.33F 5.5V Coin Cell Supercapacitor KR-5R5V334-R Memory Backup Multi

[Zhinox]

So the resistors show different values and added together they again give the total number of about 5.2 kOhm.

The super condenser holds the 4.3 volts even when the device is turned off. I replaced it with a new one. In the picture above you can still see the old broken supercapacitor. This has leaked.

If they tell me how I can measure d9, so what to adjust on the multimeter, then I can do that. This also applies to r170 🙈

[CuriousJohn]

thanks for the photos appreciated.

The servo measures ok then, good to confirm.

The Supercap is holding its charge of 4.3v when no power to the board, its ok by sounds of it. The supercap is supplied with 5v through D9 and R170 should measure 100 Ohms if you measure the resistance of it with your meter when power is off, nice little calculator for working out SMD resistors to numbers on top = the value in ohms it should be: https://www.hobby-hour.com/electronics/smdcalc.php.

As for the diode that SOT23 package contains only 1 diode as drawn in the diagram, one leg of that component is not connected to anything inside and the diode is the reason 5v dc can go to the supercap from the power supply but not go back as the diode prevents current flow the other way. A curiosu thing about diodes is they consume some volts its called forward bias of 0.7v and thats why the supercap is charged to 4.3v only from a 5v supply as 0.7v is used by the diode when its working right so it sounds ok.

Anyway to test this diode D9 with power off measure the resistance between the Anode and Cathode, then again the other way because of the way diodes work we need to look at the resistance of it in both directions. The resistance should be low when Red probe on Anode pin which is the top leg on the side with two legs, the cathode is all by itself on the other side. Resistance should measure very high with red probe on cathode and black probe on anode it stops votlage going back out the supercap to the U16 that supplies it. Be aware though that measuring resistance using a multimeter ohm meter when there is power to the board may give false results so the supercap will need to be removed or at least one leg disconnected or something. This will also provide you with the answer of if the fault goes away when the super cap is not connected. Bit of a chore i know, i dont another way maybe somebody else does...


TBH if you dont use the JRNY app and still have this fault now it would be simplest to remove the supercap and see if it works for you like that.

EDIT

Feeling brave? Show a photo of the underside of the board around the supercap area and we can see if there is short due to poor soldering or something again another chore taking it all apart again especially those Zebra strips lol. There are some fine traces around in that area and a simple solder splash could cause issues it would help to inspect it all with a magnifying glass or Jewellers loope. again consider just removing that supercap for a quick fix and to verify it is the supercap causing it.

[Zhinox]

So, I desoldered the super capacitor. The problem is exactly the same. If I connect the power and unplug the servo, the console starts and I can plug in the servo and the motor turns. If I leave the servo plugged in, the console does not start. Only when I remove the servo does the console start. But if I plug in the servo and plug in the power, nothing happens. However, if I pull the power plug, the console lights up briefly.

The picture shows the back of the console. The super capacitor was still soldered in. In the other picture I removed it.

[Zhinox]

I also measured the r170. Here I had 100 ohms.

[Zhinox]

I also measured d9. If I put red on the top foot (where there are two) and black in the middle, then I get 0.993

If I measure it the other way around, then the 1 stays (black on top and red in the middle)