No signal on D+ or D- when I plug the USB

I'm working on the vendor_pic32mz_ef_sk example to develop a USB device with a PIC32MZ2048EFH144. I'm using an EasyPIC Fusion v7 by Mikroelektronika configured as the PIC32MZ Embedded Connectivity with FPU (EF) Starter Kit for the LEDs and SWITCHs. The external clock is the same, 24MHz oscillator, and the USB bus seems to be the same too, except for two 27Ohm serie resistors on D+ and D-, and a 220Ohm resistor and a BAT43 diode in serie on VBUS:
 
EasyPIC Fusion v7 bus
[See EasyPIC_Fusion_v7_USB_Bus.png attached]
 
PIC32MZ Embedded Connectivity with FPU (EF) Starter Kit bus
[See Starter_Kit_USB_Bus.png attached]
 
I've loaded the vendor_pic32mz_ef_sk example on the PIC32MZ but when I plug the USB cable nothing appen: no USB plug sound and no signals on both D+ and D-. The project is configured as High speed device so I espect an high signal on the D+ pin for the speed negotiation. But nothing appen: I checked with oscilloscope on D+ and D- but they are always low.
I'm pretty sure that the USB_DEVICE_EVENT_POWER_DETECTED event is triggered
 

case USB_DEVICE_EVENT_POWER_DETECTED:

            /* VBUS is detected. Attach the device */
            USB_DEVICE_Attach(appData.usbDevHandle);
            Task_Led_On();  //I turn on the led to confirm the event
            break;

 
and nothing else. But if I check the USBOTG, USBCSR0 and USBCRCON registers before and after I plug the cable, I get:
 
Cable disconnected

register USBOTG: 00 00 00 80
BDEV = 1: USB is operating as 'B' device

register USBCRCON: 01 00 80 04
USBWK = 1: Connect, disconnect, or other activity on USB detected since last cleared
USBIE = 1: Enables general interrupt from USB module
USBRIE = 1: Enable remote resume from suspend Interrup

register USBCSR0: 00 00 20 00
HSEN = 1: The USB module will negotiate for Hi-Speed mode when the device is reset by the hub

 
Cable connected

register USBOTG: 00 00 00 98
BDEV = 1: USB is operating as 'B' device
VBUS = 1: Above VBUS Valid

register USBCRCON: 01 00 80 04
USBWK = 1: Connect, disconnect, or other activity on USB detected since last cleared
USBIE = 1: Enables general interrupt from USB module
USBRIE = 1: Enable remote resume from suspend Interrup

register USBCSR0: 00 00 70 00
SOFTCON = 1: The USB D+/D- lines are enabled and active
HSEN = 1: The USB module will negotiate for Hi-Speed mode when the device is reset by the hub
HSMODE = 1: Hi-Speed mode successfully negotiated during USB reset

 
The VBUS field goes to '11' then the level is valid and the HSMODE goes to '1' too, which means that Hi-Speed mode negotiation occurred!
My first thought was that the internal (?) pull-up resistors on the D+ and D- were not activated (strange, since the USB_DEVICE_Attach function in the USB_DEVICE_EVENT_POWER_DETECTED was called), so I tryed to hard connect an external pull-up resistor on D+. With that resistor I can get the plug sound from  my host PC and I can see some signals on D+ but the host can't recognize it:
 
[See USB_Error.png attached]
 
But at least is a little step forward. I think the problem is at its source.
 
Why whitout external pull-up resistor the D+ and D- lines are low? It should work without it, right?
And why the HSMODE is set while I have no negotiation signals on D+ or D-?
 
PS: I've tryed on another PIC and same problem.

MisterHemi
I don't use Harmony but know a bit about the PIC32MZ's USB
 
The internal resistors are connected to the D+ and D- lines when SOFTCON = 1
 
At which point the host (your PC)or hub should recognize that a device has attached itself to the bus and it should attempt to start the enumeration process.
 
I have also had similar issues with different boards. Some worked when using a hub in between the device and computer and others have worked fine when connected directly to the computer (in my case a MacBook Pro).
 
You can use software such as Wireshark to monitor the USB and see if it recognizes the device and if it begins enumeration.
 
Also be sure to remove the external resistors you connected as this would cause errors such as making the host recognize the device as a different speed, e.g. - low speed vs high speed, etc.

Thanks for your point but:
 
- As I shown, the SOFTCON goes to '1' after I plug the cable; so the D+ and D- should be connected to the internal pull-up resistors.
 
- I'm monitoring both D+ and D- bacause first of all, before the enumeation, I expect the speed negotiation: D+ pulled high for High/Full Speed device, D- pulled high for Low Speed device. But on the D+ and D- I have ALWAYS low level, then no speed negotiation and consequently no enumeration.
 
- I've added the external pull-up resistor only to try. And it seems work! But why, without external resistor, I have SOFTCON = 1 and D+ and D- ALWAYS low?
 
I plug the cable from board directly to PC USB port; no HUB, cause using it I have the Vbus level lower than the valid limit. Connected direct to PC I have Vbus = 4.9V .
 
I read in another post that a pull-down resistor on Vbus line solved the problem (in the Starter Kit schematic I've attached in the previous post the pull-down resistor is present!), but I've tryed it and problem occurs again. Is necessary a pull-down resistor on Vbus?

I believe it's necessary to have the pin connected or use the code example I posted to force the PIC into device mode.
 
Try that and let me know if it works.

thoraz

MisterHemi
I believe it's necessary to have the pin connected or use the code example I posted to force the PIC into device mode.
 
Try that and let me know if it works.

Today is a good day thanks to MisterHemi! grin:
 
You're right and I understood the source of my problem: on the EasyPIC Fusion I had the USBID pin (RF3) pulled down by one resistor enabled by a switch (the SW7-RF3, for the ones that have an EasyPIC board and are reading this post). Than, I disabled that pull-down resistor and all works fine.
 
I can keep the pull-down resistor active, but in that case I need to set the USBCRCON bits as MisterHemi said to override the signal on USBID pin.
 
The OTG feature, with the USBID pulled down, leads the microcontroller act like a host device and that's why the D+ and D- signals were both keep low.
 
Thanks again.

You're welcome!
 
Glad you were able to get it working!