Rehab1 Posted June 6, 2017 Author Share Posted June 6, 2017 30 minutes ago, Marty Backe said: I love the enthusiasm and knowledge sharing in this group, and how small the planet is now (imagine doing a similar collaboration in 1920). 1920...or just about anytime BI (before internet). I cannot imagine a time when there was no Spell Check! Sorry to rub it in! Link to comment Share on other sites More sharing options...
Dingfelder Posted June 7, 2017 Share Posted June 7, 2017 3 hours ago, Marty Backe said: I love the enthusiasm and knowledge sharing in this group, and how small the planet is now (imagine doing a similar collaboration in 1920). Me too, though I'm too ignorant to share in the knowledge. That leaves me just appreciating the enthusiasm! Now if you folks were discussing things on the level of rolling some dough into a ball or stirring sugar into iced tea, I could follow no problem. Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted June 7, 2017 Share Posted June 7, 2017 12 hours ago, Dingfelder said: Me too, though I'm too ignorant to share in the knowledge. That leaves me just appreciating the enthusiasm! Now if you folks were discussing things on the level of rolling some dough into a ball or stirring sugar into iced tea, I could follow no problem. yeah its amazing. we all speak different languages and it all shows up in English! Amazing! LOL. I wonder? If there are extraterrestrials among use, would they join the chat forums? Perfect way to interact and not be discovered. They could use bad grammar and spelling and now one would know! Link to comment Share on other sites More sharing options...
Rehab1 Posted June 7, 2017 Author Share Posted June 7, 2017 2 hours ago, Carlos E Rodriguez said: They could use bad grammar and spelling and now one would know! Oh many forum members would know, especially the spelling aspect! Just ask @Marty Backe about his recent episode! Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted June 8, 2017 Share Posted June 8, 2017 10 hours ago, Rehab1 said: Oh many forum members would know, especially the spelling aspect! Just ask @Marty Backe about his recent episode! So you are saying @Marty Backe is an extraterrestrial? Link to comment Share on other sites More sharing options...
FULspeed Posted June 8, 2017 Share Posted June 8, 2017 On 30/05/2017 at 9:36 PM, Rehab1 said: Has anyone tried installing a fan on their ACM obtaining voltage from the socket on the main board depicted in photo #2? Hi Rehab1, I was looking your picture and I think that the problem comes from the socket, It's not the right one... I confirm that in my ACM 16 1300, the added fan works very well and that it start when the board temp reach the 35 Celsius (95 F) (checked with an hairdryer) Try the near socket, that one in vertical position, (see photo) it should work. ps. It's my english that's terrible, i'm not an alien... Link to comment Share on other sites More sharing options...
Hansolo Posted June 8, 2017 Share Posted June 8, 2017 Like this (MS3 1600) the connector used is surrounded in yellow Link to comment Share on other sites More sharing options...
Rehab1 Posted June 8, 2017 Author Share Posted June 8, 2017 1 hour ago, FULspeed said: Hi Rehab1, I was looking your picture and I think that the problem comes from the socket, It's not the right one... I confirm that in my ACM 16 1300, the added fan works very well and that it start when the board temp reach the 35 Celsius (95 F) (checked with an hairdryer) Try the near socket, that one in vertical position, (see photo) it should work. ps. It's my english that's terrible, i'm not an alien... 1 hour ago, Hansolo said: Like this (MS3 1600) the connector used is surrounded in yellow Many thanks guys! UGH.....I just wired up my fan and drilled the hole in the side panel for the switch. Link to comment Share on other sites More sharing options...
Rehab1 Posted June 8, 2017 Author Share Posted June 8, 2017 1 hour ago, FULspeed said: ps. It's my english that's terrible, i'm not an alien... Link to comment Share on other sites More sharing options...
esaj Posted June 8, 2017 Share Posted June 8, 2017 I poked around the board more, but haven't been able to figure out what exactly is wrong with it. I have a few possible clues though. This seems to be pretty much the step-down stage of the board. The weird thing is, there actually seems to be three different step-down chips, don't know if they're completely separate or (partially) in series yet, and a 3.3V linear regulator. The DK1203 is next to the large inductor/coil-block, XL7005A and SM7025 are to the side, with their own smaller inductors & capacitors next to them (just outside the picture on the left). The two smaller images I've set to the side have a couple of things that I'm focusing on next (maybe tonight, or later, don't know yet). The lower one with the marking "where does this go?" has a trace that at first glance appeared to go the the base (marked as "B") -pin of the MMBT5551 NPN transistor, but upon further measurement it doesn't. Instead, it seems that the trace leaving from there goes to the other side of the transistor and then is taken somewhere else through a via (where it goes, I don't know yet, at least it doesn't appear go to ground). The upper small picture is from just above the same NPN transistor. There's two 2.2K resistors there, other going to the ground plane next to the resistors on the same layer, and the other has a trace that seems to run underneat the XL7005A-chip. The weird thing is, it appears to be shorted directly to ground. I think this is some indication of the fault on the board, as it wouldn't make sense for them to run a long trace just to take the resistor to ground, since there's a ground plane right next to it, and probably another one on the other side of the board, so if they wanted to ground it, they could have just connected it to the same ground plane as the resistor next to it, or run a via (assuming there's space in the layers in-between) to the ground plane-layer. The thing is, it would seem to go to the EN (enable) -pin of the XL7005A, but the EN -pin is not shorted to ground. I think that this trace runs to something that enables the step-downs and turns on the entire wheel when the power-button is pressed, but I need to measure it further (both with and without power) to make more sense of it. Luckily, the power to the probably shouldn't be shorted to ground but still is -resistor seems to come through high resistance, so there isn't much current flowing, thus it likely shouldn't make more damage even when the board's powered. The circuit up to that point looks like this (barring any mistakes): EDIT: Forgot that I drew this when I still thought that that one trace from the 2.2K -resistors runs to the base of the NPN, so this is wrong... R6 doesn't go to the base of the Q2. Link to comment Share on other sites More sharing options...
Rehab1 Posted June 8, 2017 Author Share Posted June 8, 2017 3 hours ago, esaj said: I poked around the board more, but haven't been able to figure out what exactly is wrong with it. I have a few possible clues though. Strong stuff Esaj! Thanks! I am certain you locate the burned out components. Finding the weak link on the board when the electrical arc occurred may be challenging but you are really making progress! Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted June 9, 2017 Share Posted June 9, 2017 On 6/8/2017 at 1:00 PM, esaj said: I poked around the board more, but haven't been able to figure out what exactly is wrong with it. I have a few possible clues though. This seems to be pretty much the step-down stage of the board. The weird thing is, there actually seems to be three different step-down chips, don't know if they're completely separate or (partially) in series yet, and a 3.3V linear regulator. The DK1203 is next to the large inductor/coil-block, XL7005A and SM7025 are to the side, with their own smaller inductors & capacitors next to them (just outside the picture on the left). The two smaller images I've set to the side have a couple of things that I'm focusing on next (maybe tonight, or later, don't know yet). The lower one with the marking "where does this go?" has a trace that at first glance appeared to go the the base (marked as "B") -pin of the MMBT5551 NPN transistor, but upon further measurement it doesn't. Instead, it seems that the trace leaving from there goes to the other side of the transistor and then is taken somewhere else through a via (where it goes, I don't know yet, at least it doesn't appear go to ground). The upper small picture is from just above the same NPN transistor. There's two 2.2K resistors there, other going to the ground plane next to the resistors on the same layer, and the other has a trace that seems to run underneat the XL7005A-chip. The weird thing is, it appears to be shorted directly to ground. I think this is some indication of the fault on the board, as it wouldn't make sense for them to run a long trace just to take the resistor to ground, since there's a ground plane right next to it, and probably another one on the other side of the board, so if they wanted to ground it, they could have just connected it to the same ground plane as the resistor next to it, or run a via (assuming there's space in the layers in-between) to the ground plane-layer. The thing is, it would seem to go to the EN (enable) -pin of the XL7005A, but the EN -pin is not shorted to ground. I think that this trace runs to something that enables the step-downs and turns on the entire wheel when the power-button is pressed, but I need to measure it further (both with and without power) to make more sense of it. Luckily, the power to the probably shouldn't be shorted to ground but still is -resistor seems to come through high resistance, so there isn't much current flowing, thus it likely shouldn't make more damage even when the board's powered. The circuit up to that point looks like this (barring any mistakes): EDIT: Forgot that I drew this when I still thought that that one trace from the 2.2K -resistors runs to the base of the NPN, so this is wrong... R6 doesn't go to the base of the Q2. @esaj. Considering the event occurred at the battery connection piece to the board, maybe poke at the first level of interfaces in the battery input circuit on the board. If you are lucky you will find the dead chip on the first interface chips that look directly at the battery volts in the board. If not then your intuition is next. Link to comment Share on other sites More sharing options...
esaj Posted June 9, 2017 Share Posted June 9, 2017 I unscrewed the board from the heatsink to find out that it's not multilayered, but just double-sided. With a couple of photos, and some resizing / mirroring / tilting the images could be aligned to see where the vias go. I can adjust the opacity or change the layers in image editing, so it's pretty easy to follow, so I now know where the "weird" vias go (PA14 & PA13 of the STM32) and what else is connected to the collector of the PNP. The thing is, the VDD's and VSS's of the STM32 are both on ground, so it looks like the it could be the 3.3V regulator, the bypass caps or the STM32 itself. I've left the board soaking in IPA to get the coating off, next step is removing the 3.3V regulator, then testing it in isolation to see if it's still working, and if it is, I'll solder some wires to the board to power the 3.3V side to see if the MCU starts to heat up or if it's the caps. If it's the 3.3V regulator, I have those exact same regulators in the same package. If it's the MCU, then there's nothing really left to do, except maybe strip the board for parts Oh, another weird thing I found on the powered board: the larger cap + diode near the DK1203 and the larger-package (1206?) resistor seem to form a voltage doubler (about 140V there with 75V input). No idea why yet. I'll see if I get around to work on it tonight or later, just got back from a 50km ride with a friend a while ago. Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted June 9, 2017 Share Posted June 9, 2017 5 minutes ago, esaj said: I unscrewed the board from the heatsink to find out that it's not multilayered, but just double-sided. With a couple of photos, and some resizing / mirroring / tilting the images could be aligned to see where the vias go. I can adjust the opacity or change the layers in image editing, so it's pretty easy to follow, so I now know where the "weird" vias go (PA14 & PA13 of the STM32) and what else is connected to the collector of the PNP. The thing is, the VDD's and VSS's of the STM32 are both on ground, so it looks like the it could be the 3.3V regulator, the bypass caps or the STM32 itself. I've left the board soaking in IPA to get the coating off, next step is removing the 3.3V regulator, then testing it in isolation to see if it's still working, and if it is, I'll solder some wires to the board to power the 3.3V side to see if the MCU starts to heat up or if it's the caps. If it's the MCU, then there's nothing really left to do, except maybe strip the board for parts Oh, another weird thing I found on the powered board: the larger cap + diode near the DK1203 and the larger-package (1206?) resistor seem to form a voltage doubler (about 140V there with 75V input). No idea why yet. I'll see if I get around to work on it tonight or later, just got back from a 50km ride with a friend a while ago. I am voting the 3.3 voltage regulator died. Link to comment Share on other sites More sharing options...
Rehab1 Posted June 9, 2017 Author Share Posted June 9, 2017 22 minutes ago, esaj said: Oh, another weird thing I found on the powered board: the larger cap + diode near the DK1203 and the larger-package (1206?) resistor seem to form a voltage doubler (about 140V there with 75V input). No idea why yet. Wow...That is a great deal of voltage. The 2nd replacement board from China arrived today after 6 weeks. Thanks to @Jason McNeil the other replacement took 2 days. If you need any static readings of the easy accessible components (without a tank dump) I would be more than happy to provide them. 23 minutes ago, Carlos E Rodriguez said: I'll see if I get around to work on it tonight or later, just got back from a 50km ride with a friend a while ago. Enjoy your well deserved ride! 26 minutes ago, Carlos E Rodriguez said: I am voting the 3.3 voltage regulator died. This could be an interesting little contest! Link to comment Share on other sites More sharing options...
esaj Posted June 9, 2017 Share Posted June 9, 2017 9 minutes ago, Rehab1 said: Wow...That is a great deal of voltage. The 2nd replacement board from China arrived today after 6 weeks. Thanks to @Jason McNeil the other replacement took 2 days. If you need any static readings of the easy accessible components (without a tank dump) I would be more than happy to provide them. Thanks, I'll let you know if I come up with of something that would need such. 9 minutes ago, Rehab1 said: Enjoy your well deserved ride! Thanks, we already got back a while ago. With almost 50km (49.xx, measured from city zoning map tools) ride with lots of uphills, still got 43% / 57.0V left... I'm pretty sure it could carry me up to 70-80km, but probably with serious limp-mode towards the end. 9 minutes ago, Rehab1 said: This could be an interesting little contest! If the 3.3V regulator's blown, it could have taken the MCU with it... the absolute maximum voltage for that is only 4V, somewhat higher for the 5V tolerant pins. Probably won't start poking around it again until tomorrow or Sunday. Link to comment Share on other sites More sharing options...
Rehab1 Posted June 10, 2017 Author Share Posted June 10, 2017 Thanks to @Marty Backe and his GW wish list I purchased a bluetooth amp and speakers so my ACM can cruise with music. Unfortunately there is no long any room for a speaker on the mainboard side of the wheel. The opposite compartment has plenty of room for both the amp and one speaker. The power for the amp will come from the 5w usb connection. Link to comment Share on other sites More sharing options...
Marty Backe Posted June 10, 2017 Share Posted June 10, 2017 47 minutes ago, Rehab1 said: Thanks to @Marty Backe and his GW wish list I purchased a bluetooth amp and speakers so my ACM can cruise with music. Unfortunately there is no long any room for a speaker on the mainboard side of the wheel. The opposite compartment has plenty of room for both the amp and one speaker. The power for the amp will come from the 5w usb connection. I can't wait to see how that turns out. It's one mod that I may consider. Will have to hear what kind of volume you get out of it. Link to comment Share on other sites More sharing options...
Rehab1 Posted June 10, 2017 Author Share Posted June 10, 2017 16 minutes ago, Marty Backe said: I can't wait to see how that turns out. It's one mod that I may consider. Will have to hear what kind of volume you get out of it. I plan to test the system out before installation. If it works to my satisfaction then I will get my drill out again! You would have an easy time installing 2 speakers with the ample room afforded in the side compartments. Sometimes I wonder if installing the cooling fan was worth the effort. Oh well it was a fun mod and I can always buy a new side panel if there is zero fruition. Link to comment Share on other sites More sharing options...
SuperSport Posted June 10, 2017 Share Posted June 10, 2017 10 minutes ago, Rehab1 said: Oh well it was a fun mod and I can always buy a new side panel if there is zero fruition. Or replace the fan w a speaker. Link to comment Share on other sites More sharing options...
Rehab1 Posted June 10, 2017 Author Share Posted June 10, 2017 5 minutes ago, SuperSport said: Or replace the fan w a speaker. I could do that but I still would have a bazillion tiny air intake holes to contend with! Link to comment Share on other sites More sharing options...
Carlos E Rodriguez Posted June 10, 2017 Share Posted June 10, 2017 1 minute ago, Rehab1 said: I could do that but I still would have a bazillion tiny air intake holes to contend with! Just fill them with the plastic melt that I posted before Link to comment Share on other sites More sharing options...
Rehab1 Posted June 10, 2017 Author Share Posted June 10, 2017 13 minutes ago, Carlos E Rodriguez said: Just fill them with the plastic melt that I posted before That is a great method! Jason had a set of panels in stock that I just ordered. With an extra set of panels it might be fun to play around adding ground effect customizations. Link to comment Share on other sites More sharing options...
SuperSport Posted June 10, 2017 Share Posted June 10, 2017 1 hour ago, Rehab1 said: I could do that but I still would have a bazillion tiny air intake holes to contend with! Yeah, I hadn't caught that. I guess you could leave them to improve the bass response? I'm not a sound guy, so I really don't know what I'm talking about. Probably heard it in a commercial somewhere. ? Link to comment Share on other sites More sharing options...
esaj Posted June 11, 2017 Share Posted June 11, 2017 The coating didn't come off cleanly in the soaking after all, while the IPA makes the coating soft and turns it into a gel-like consistency, it still needs mechanical brushing to come off. Maybe acetone might work better, but it might also attack some of the plastics, so probably won't try it. I got enough exposed to work on it, although I could smell the burning coating while I desoldered the 3.3V regulator. The regulator's fine, even after I had to mangle the legs a bit to get it off, still regulating as it's supposed to when tested separately. But the STM32's dead, after a while of being powered, it started to evaporate IPA fast and heated up enough to start to burn my finger (did the usual "power with 3.3V / limited current, see if it draws too much current and what heats up"). So probably no hope to fix the board, just changing the chip won't help without the firmware burned onto it. Can't really tell what burned it, ie. whether a higher voltage spike got through to the chip or something, but PA13 is shorted to ground, as well as the power inputs. This is as far as I went with the power-input side. I'll see if I can sometime try to draw schematics for the board, while it has a lot more components than the Firewheel-board, it doesn't have internal layers, which would make things easier. Then again, it also has no designators for any components, so matching the schematic to the board might be tricky and make it less useful. For anyone interested, here're the chips on the board: DK1203 AC-DC Switch Mode Power Controlling IC http://grupoautcomp.com.br/wp-content/uploads/2016/11/Specification-IC-DK1203.pdf (I guess the "AC" here means it can withstand high voltages, the typical application sample in the datasheets shows a rectification bridge before the chip) XL7005A 0.4A 150KHz 80V Buck DC to DC Converter http://www.xlsemi.com/datasheet/XL7005A datasheet-English.pdf SM7025 Some sort of buck/boost controller-chip, haven't found an English datasheet https://wenku.baidu.com/view/0d41c4370b1c59eef8c7b4c5.html AMS1117-3.3 3.3V linear regulator http://www.advanced-monolithic.com/pdf/ds1117.pdf PC817 optoisolator / photocoupler http://www.sharp-world.com/products/device-china/lineup/data/pdf/datasheet/pc817xnnsz_e.pdf STM32 F103C8T6 MCU http://www.st.com/content/ccc/resource/technical/document/datasheet/33/d4/6f/1d/df/0b/4c/6d/CD00161566.pdf/files/CD00161566.pdf/jcr:content/translations/en.CD00161566.pdf ACS709LLFTR-35BB-T 75A Current sensor on two phases (also, looks like a 1 milliohm bypass resistor in parallel with them) http://www.allegromicro.com/~/media/Files/Datasheets/ACS709-Datasheet.ashx IR2184S Half-bridge drivers x 3 http://www.infineon.com/dgdl/ir2184.pdf?fileId=5546d462533600a4015355c955e616d4 MPU-6050 Gyro/accelerometer https://www.invensense.com/wp-content/uploads/2015/02/MPU-6000-Datasheet1.pdf The mosfets are IRFB4110's (TO-220's), 12 in total, 2 in parallel (there's no resistors or ferrites between the gates of parallel mosfets) http://www.infineon.com/dgdl/irfb4110pbf.pdf?fileId=5546d462533600a401535615a9571e0b There's also a number of discrete transistor around the board. I can upload more pics of the board if someone needs them. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.