New PID needed for APC UPS

[jspitael]

Hi,

I work for Schneider Electric and we will be introducing a new APC UPS that will use a VID of 0x051D and a PID of 0x0005. We request that you add those values to https://raw.githubusercontent.com/networkupstools/nut/master/scripts/upower/95-upower-hid.hwdb.

If other information is required to support this UPS in NUT, please let me know.

Thanks,
Jim

[jimklimov]

Cheers, you're quite on time - a NUT v2.8.4 is planned for release before the end of this week.

Which driver would support the device, usbhid-ups? (the file you referred to is one of many generated from what the drivers declare)

Gotta make sure of this, as lately (since 2010, or so "they" say) APC sadly seems to deprecate standard USB HID support in favour of Modbus, Microlink or other protocols. Can you somehow comment on that, or at least pass on feedback that this is annoying? And makes it harder than it could be to provide quality drivers based on educated guesses, trial and error... Vendor cooperation (docs, specs, perhaps even code PRs) would be very much appreciated!

Thanks,
Jim

[jimklimov]

If it supports USB HID, please also check the apc_usb_device_table[] at

nut/drivers/apc-hid.c

Lines 89 to 104 in a3e0664

	/* USB IDs device table */
	static usb_device_id_t apc_usb_device_table[] = {
	/* APC AP9584 Serial->USB kit */
	{ USB_DEVICE(APC_VENDORID, 0x0000), NULL },
	/* various models */
	{ USB_DEVICE(APC_VENDORID, 0x0002), general_apc_check },
	/* various 5G models */
	{ USB_DEVICE(APC_VENDORID, 0x0003), disable_interrupt_pipe },
	/* APC Smart UPS 1000 with latest firmware 04.3
	* seems to have bumped the productid from 3 to 4
	* See https://github.com/networkupstools/nut/issues/1429 */
	{ USB_DEVICE(APC_VENDORID, 0x0004), disable_interrupt_pipe },
	/* Terminating entry */
	{ 0, 0, NULL }
	};

- most other series needed this or that tweak method to be referenced. Is one of those applicable here?

Otherwise, a likely suitable driver is https://github.com/networkupstools/nut/blob/master/drivers/apc_modbus.c but it has certain stability issues (largely due to libmodbus, I think, but for end-users it is same - driver either works or not overall).

[jimklimov]

Per offline discussion, answering the questions above would need some more time so pushing to next NUT release.

[jimklimov]

@jspitael : I'm getting back from the offline vacation soon; are there any news on your side? (Which of the protocol drivers would the new USB IDs best apply to?)

[jspitael]

Hi Jim, UPSs with PID of 0x0005 will support the HID power summary. Exactly which usages are still not finalized. Is that sufficient for now? Regards, Jim

…

________________________________ From: Jim Klimov ***@***.***> Sent: Tuesday, September 2, 2025 5:25 AM To: networkupstools/nut ***@***.***> Cc: jspitael ***@***.***>; Mention ***@***.***> Subject: Re: [networkupstools/nut] New PID needed for APC UPS (Issue #3047) [https://avatars.githubusercontent.com/u/1636756?s=20&v=4]jimklimov left a comment (networkupstools/nut#3047)<#3047 (comment)> @jspitael<https://github.com/jspitael> : I'm getting back from the offline vacation soon; are there any news on your side? (Which of the protocol drivers would the new USB IDs best apply to?) — Reply to this email directly, view it on GitHub<#3047 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BUD2BQ576OBSZM3VHMDF7533QVPAXAVCNFSM6AAAAACDKWPLXGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENBUGUZDGMJWGU>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[jimklimov]

I guess so, thanks. Eventually, check with drivers/apc-hid.c when the device is ready, to see if needed data points are handled (mapping NUT concept names per docs/nut-names.txt to USB paths).

FWIW, I'm also working on a solution for issue #3082 to help tracking down usages that are not mapped (seen in the report, not referenced by run-time walk of the mapping table).

[jimklimov]

Commit posted to a relevant PR that's in progress. So far it is a sort of placeholder - notably without referencing a tweak method (no idea if any would be needed for the new lineup). For examples of earlier tweaks see e.g.

nut/drivers/apc-hid.c

Lines 40 to 104 in a3e0664

	/* Tweaks */
	static char * tweak_max_report[] = {
	/* Back-UPS ES 700 does NOT overflow. */
	/* Back-UPS ES 725 does NOT overflow. */
	/* Back-UPS ES 525 overflows on ReportID 0x0c
	(UPS.PowerSummary.RemainingCapacity). */
	"Back-UPS ES 525",
	/* Back-UPS CS 650 overflows on ReportID 0x46 */
	"Back-UPS CS",
	NULL};
	/* Don't use interrupt pipe on 5G models (used by proprietary protocol) */
	static void *disable_interrupt_pipe(USBDevice_t *device)
	{
	NUT_UNUSED_VARIABLE(device);
	if (use_interrupt_pipe == TRUE) {
	/* FIXME? Suggest data from "device" to help the setup below? */
	upslogx(LOG_INFO, "interrupt pipe disabled (add 'pollonly' flag to 'ups.conf' to get rid of this message)");
	use_interrupt_pipe= FALSE;
	}
	return NULL;
	}
	/* Some models need special tweaks */
	static void *general_apc_check(USBDevice_t *device)
	{
	int i = 0;
	if (!device->Product) {
	upslogx(LOG_WARNING, "device->Product is NULL so it is not possible to determine whether to activate max_report_size workaround");
	return NULL;
	}
	/* Some models of Back-UPS overflow on some ReportID.
	* This results in some data not being exposed and IO errors on
	* WIN32, causing endless reconnection or driver's failure */
	while( tweak_max_report[i] != NULL ) {
	if(!strncmp(device->Product, tweak_max_report[i],
	strlen(tweak_max_report[i]))) {
	max_report_size = 1;
	return NULL;
	}
	i++;
	}
	return NULL;
	}
	/* USB IDs device table */
	static usb_device_id_t apc_usb_device_table[] = {
	/* APC AP9584 Serial->USB kit */
	{ USB_DEVICE(APC_VENDORID, 0x0000), NULL },
	/* various models */
	{ USB_DEVICE(APC_VENDORID, 0x0002), general_apc_check },
	/* various 5G models */
	{ USB_DEVICE(APC_VENDORID, 0x0003), disable_interrupt_pipe },
	/* APC Smart UPS 1000 with latest firmware 04.3
	* seems to have bumped the productid from 3 to 4
	* See https://github.com/networkupstools/nut/issues/1429 */
	{ USB_DEVICE(APC_VENDORID, 0x0004), disable_interrupt_pipe },
	/* Terminating entry */
	{ 0, 0, NULL }
	};