MC2100 treadmill motor control circuit

Thread Starter

domingojordan

Joined Aug 11, 2014
17
I am interested in the top part of this circuit... I will like to reprogram the embede micro controller that it is not shown in this diagram. If any one knows how can I program a cypress psoc microcontroller or if you know the author of this circuitry, I will apriciate if you can share it with us. I need to alter the way the motor behaves when it speed is manage. I want to thank Mike ml and the others for answers some questions that i had.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
I don't have the authors details, but I am not sure if that will help if you want to program your own μpr.
The reverse engineering was done for the purpose of trouble shooting the factory version.
Max.
 

Thread Starter

domingojordan

Joined Aug 11, 2014
17
Max head room, In the reverse engineering circuit for the MC2100 treadmill motor controller, I believe it was done to reprogram the controller. The eng set up a description of the 20 new pins and its new use. They are shown in the circuit shown in black at the bottom right corner. This is the reason that I would like to get in contact with him. I need to do the same thing.

My question to you is how do they limit the current that goes to the motor? I tested my motor in the circuit, eliminando the mosfet. My motor works but it gets really hot and it is not happy. I believe that it draw to mouch current. When I run it with the mosfet it runs fine. I can control the speed of my motor. Where an how does this circuit limits the current that the mosfet provides to the motor? I am talking about the circuit and motor at the top.
I have this controller and it works fine with my motor but when I chage the speed it take some time to to set to the new speed which is normal in the treadmill industry but is not for my project. I need to ramp up faster or ramp dow faster wherever is the case. Thank you very much for your time.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
For what you need, I would suggest either looking for a KB or Baldor drive, either SCR bridge or PWM style, or just designing your own PWM drive from scratch.
Max.
 

IamJatinah

Joined Oct 22, 2014
136
Hi Folks, if looking for adaptive speed controls for PWM driven motors, message me. I have built and adapted many styles of controllers available for lathes and other milling applications. :eek:)
 

vermejc

Joined Oct 17, 2015
2
20151015_111955.jpg 20151015_112011.jpg Hi
I have a MC2100ELS controle border
The motor move with 12V baterie, so it's ok
I have 138 ACV at sencor motor movement out-put. (HD5) I suppose this is the problem.
Could you please send me the electrical shceme
Thanks in advance
 
I have a MC2100ELS board but it does not match the diagram above. It has a blown capacitor on the power input and what looks like a small rectifier also blown. I am trying to identify the rectifier, marked GT1 on the board. I have not bothered to trace it yet (just lazy I guess), so any help will be gratefully received.
 

IamJatinah

Joined Oct 22, 2014
136
Hi All. Yep, this pictured unit is a new version of the MC2100. The shown schematic won't be correct, and the small uP on the board may not be reprogrammable depending on it's authors coding and fuse availability in most uP's. I haven't seen this unit yet to redraw it, but I can tell you folks a few things on here that have changed from the original versions. Icon now has dual input storage capacitors when they used to have one, this increases the current drive and stability of the controller and it's regulation. HD5 should be a roller feedback connector, and if you have 138ACV on these two pins, there is something very wrong going on, as this connection simply "shorts" a logic high level each time the front roller magnet mates with the reed switch or hall effect detector.
The three pin connector should be your elevation feedback connection, and finally the longer connector should be command lines for lift up, lift down, enable(safety enable if incorporated), and a speed command. Originally these command lines were all ttl-logic in an analog format, and I assume the same on this design.
For those engineers considering digital controls for treadmills, I have a simple message for you, don't. Digital control of treadmills has been a bit of a fight for manufacturers and response times for many situations lag the analog speeds, believe it or not ;o) Have a great day! Ooops... your Diode marked GT1 crosses to the Rohm BC848B diode, shown to be a small signal diode? If this is blown there may be other trouble. Fitness Remedy repairs most fitness electronics if needed...
 
Last edited:

zombywoof

Joined Jan 11, 2016
2
Hi All, (new here),
I have been trying to figure out just how this controller works in it's native treadmill environment. The one I have is the one pictured above on a ProForm 720 running off 230V in the UK. There seem to be a number of different designs of this MC-2100 board around, with little in the type number to indicate very major differences in design.
2 things are evident (on mine at least)
1. The reed relay ('feedback'?) speed sensor doesn't appear to do very much, although in one of the test modes on the treadmill control panel it *is* used to display the 'real' speed vs the selected target speed. However, it doesn't appear to be used as a feedback element in the control of the belt in use and there seems to be evidence of some other feedback mechanism (motor pulses / current?), albeit very coarse. Using the pot on the board to set the actual vs target speed seems to work for a while, but soon the actual speed drifts away from the target, and the pot requires re-adjustment. In summary, there does not appear to be an overall servo mechanism, and the speed control appears to be more or less open loop. This means that both speed and distance can be significantly incorrect which, for a treadmill, is - well - silly.
2. the board is fed from rectified mains (no transformer). CAREFUL!! The use of optos means there are at least 2 different GNDs - there is the EARTH GND (seen on the photo above in the bottom left corner) which IS connected to the mains supply earth, and another GND which actually sits at about -120V which is the -ve of the rectified mains. you can see another 'GND' label near top of the power resistor R28 in the photo. The HD5 connector is referenced to the -120V GND, so although the manufacturer's doc mentions it's a 5V logic signal, they failed to mention that it's sitting at a nominal -120V below EARTH GND. You can see this on the schematic at the top where the 'floating 5V' is 5V above the rectified mains. I don't know if this topology is the same on 120V AC supplied units. Needless to say - DO NOT connect the 2 GNDs together. All you will get is a noise emitting rectifier (once!).
It's interesting how complex this board appears to be for what appears to be a pretty basic control strategy...
 

IamJatinah

Joined Oct 22, 2014
136
Hi Zombywoof, nice diagnosis, you hit some good points. No "transformer" on the board as well as more than one opto indicates multiple grounds and not only that, but the Motor Power Rail uses an "Off-Line" rectified source with a HOT ground. Do NOT cross grounds when working on these, and beware most of that MC2100 logic control is attached to this HOT ground. The regulation for these treadmills are quite loose, especially on Icon branded units. I haven't dug out the redrawn schemo but I recall this feedback issue and I think they rely on lookup tables inside the small uP control IC for regulations of selected PWM (speed command) to a current window, which are compared and adjusted for regulation. That speed sensor plays a role in calibrations, but is still loose and moreso relies on the internal lookup tables of the programmed IC controlling the MCB.
 

zombywoof

Joined Jan 11, 2016
2
It’s all a bit bizarre, really. If I were designing one of these I would have put in some closed loop feedback with a time constant of a number of seconds so that it averaged out over a number of foot-strikes. Alternatively, use something like a calibration sequence at power-up such that the MCU uses the input from the speed sensor and adjusts the drive to the motor accordingly until the next session. All the circuitry and MCU smarts are there to do all this. It almost looks as though someone designed the controller ages ago and subsequent engineers have somehow dropped/ignored the (original?) concept of closed loop control. I get the impression that the motor is heavily overdriven to achieve some kind of load-related stability. To rely on this, set by a pot (with maybe a current draw sensor) – even with MCU tables – seems a bit like a student project before the ‘feedback 101’ lesson. I tried disconnecting the speed sensor and apart from it showing zero in the ‘calibration’ mode, it had no discernible effect in normal operation.

For us fitness-interested people who are attempting to observe single-figure percentage variations (for example, a minute or two over an hour – definitely significant), to have the treadmill vary over 3% from one session to the next actually makes a total mockery of any of the stats shown on-screen – the cumulative distance is calculated from target speed * time, meaning that both actual speed and actual distance are incorrect. Temperature, humidity and the user’s mass appear to have significant effects on the accuracy that is not compensated for. In the laughable ‘calibration’ mode the pot is set such that at max target speed the actual speed is within 2%, and all of the other speeds are extrapolated from that. Also, interestingly, the treadmill shuts down suddenly after 1 hour 40 mins of continuous operation – maybe as a safety or overheat protection? – and all the stats are reset to zero. To restart it you just pull out the key and reinsert it – no cool-down protection time is required. Another incomplete feature?

It just seems like they are missing the whole point of speed control...

Rant over.
 

IamJatinah

Joined Oct 22, 2014
136
Rant understood....I did not say all were designed this way, but I did say the Icon line is the lowest design-to-market folks I have seen. Fitness manufacturers tend to have things made in China, this opens many many problems as they do not adhere to US or even IEA/IEEE processes and are driven by cheapest is best, balancing performance, usually not very well.
Good fast regulation is achieved via a motor shaft encoder, or "wheel of death" to get super fast regulations ;o)
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
, you hit some good points. No "transformer" on the board as well as more than one opto indicates multiple grounds and not only that, .
The popular US manuf. of DC motor drives, KB, traditionally use transformer-less units, The low voltage in some boards are derived from a zener circuit. Some models do however have an control isolation transformer.
Also many of the US made DC and BLDC servo drives use a similar technique.
Max.
 

IamJatinah

Joined Oct 22, 2014
136
We must remember the Isolation needed between these motor drive rails, either AC or DC formats, and the control power and signals. The control signals must be isolated to above 1kv from the hot grounds residing on the drive rails. If you find a transformerless drive that also supplies control voltages I'd be interested in knowing who to stay clear of. If this isolation is not incorporated, the operators dials, switches, and other touchables must be isolated.
 

narpam2

Joined Dec 28, 2016
6
Hello everyone. I have a problem with the motor control circuit of the MC2100E Rev C treadmill. It had a short circuit in the IGBT and the R45 and R46 resistors burned out. When I put it into operation, I used the differential switch. I changed the IGBT, but the problem is that I do not know the value of resistors R45 and R46. Could someone tell me what their value is?
 
Top