O-1814/GRC-206 Rubidium Frequency Standard

The GRC-206 Pacer Speak system optionally uses a master frequency standard, the O-1814. This system is composed a MT-6250 rack holding radios for HF, VHF and UHF all of which are linked to a common control panel. The control panel can be located at the end of a 3 km fiber optic cable. Two independent control panels can be used if the operators don't try to use the same radio. The URC-113 is the HF radio system that uses a PRC-104 but with a higher power AM-71148 RF amp.

The front panel has pushbuttons for Receive and Send Time Of Day which means it has an internal clock so it's both a frequency and a time standard. Precision time is needed for modern frequency hopping military radios. The RT-1319 UHF radio needs the TOD as well as the SINCGARS VHF low band radios like the RT-1439 or RT-1523.

I think the reason for using a Rb frequency standard in this system while you don't see them in Army SINCGARS installations is the need for UTC accurate TOD for this application. In the army application the net controller can set the TOD and download that time into all the netted radios. Even if the net control time is wrong all the radios will interoperate. But when an aircraft is inbound from someplace far away (the TOD on the aircraft was set at it's home base) the TOD of the aircraft and the Pacer Speak system need to be the same for the radios to interoperate.

The inspection stamp on the Rubidium standard is 8311.16 which is week 11 of 1983. This predates the Gulf War and the introduction of GPS into large scale military operations.
STANAG 4430 was started in 1988 so was not the source of the TOD protocol. But the elements of 4430 look like a good match, they are:

1 PPS signal
Standard Time Message (STM)
Extended Have Quick Message (XHQ)
an optional 5 MHz signal.
bi-directional interface is supported. An equipment can both receive and send the TOD.

The TSQ-198 Tactical Terminal Control System (TTCS) photo on the FAS web page does not show the O-1814 installed in the rack. So the later versions probably replaced this reference with a GPS time fix?

An email from a GRC-206 user said [my comment]:

"The premise was this. Lets say the radio system was about to go down, for what ever reason (like the generator was about to die or whatever). Theory was that you could take the HQ time from your UHF radio and pump it across and the RFO [O-1814 Reference Frequency Oscillator] would hold onto that time while you got the radio system back online. That was the theory anyway.

Our experience was that it would hold a valid time for about 15 minutes and then die. My watch kept better time. Sooooo, what we did instead was just take the RFO out and put it in the conex and use the hole to mount either a TV set, paperback books or some other such stuff. The RFO was a total waste of our time and effort as far as we could tell." -Charlie-

As for the RFO, it was crap from the day we got it.  As big as it was it couldn't keep accurate time for more than 15 minutes.  Supposedly it was some big scary crystal time system in there.  My PRC-117F keeps very accurate time with its keep alive and keep it for days.  Something was just never right with the RFOs
-Charlie-

This suggests it was not a backup battery problem but rather a functional problem.  I'll check it when the Polaris GPS receiver arrives to allow loading the Have Quick time into the RFO.

This unit would keep very accurate time when running but when the external power fails it falls back on it's internal batteries. Charlie's comment ...15 minutes and then die." seems to be saying it quit working, not that it had the wrong time. More below in the Backup Battery section.

The GRC-206(V)3 system is the only one of the first four versions (V)1, (V)2 and (V)4 did not use the O-1814) that used the O-1814. (V)1 had crypto but not Have Quick (ECCM) and (V)2 and (V)4 did not have either crypto or Have Quick.

The O-1814 was designed specificially for the GRC-206 system. So the presence of a 5 MHz output seems to suggest it's there to feed the 5 MHz reference frequency input on the AM-7148 that sends it to the synthizer in the RT-1209 to put the HF radio exactly on frequency. The documents that are specific to the RFO aspect of the system don't show any cable between the RFO and AM-7148. Why? Was it an oversight, or not needed, or didn't work, or . . . If you know please tell me?

The 5 MHz output can be phase locked to the optional 5 MHz input. The testing procedures for the RFO contain checks to see that this works. But the only input source that would make sense would be a Cesium Frequency standard, like the FTS4060. This may have been part of a way to confirm the quality of the Rubidium standard in the O-1814 or some other unknown reason.

The O-1814 is supplied DC power when there is DC power on either the Vehicle or Aux input connectors to the GRC-206 system. Turning off the master switch on the Power Distribution Unit does not turn off power to the O-1814.

WANTED

Front Panel Controls, Indicators & Connectors

Label

Inside

Guesses

Connectors

Operation

The POWER connector is wired:

Pin	Function
A	Chassis Gnd
B	+ 20 to 32 VDC
C	DC return

The Battery contains 9 Cyclon Lead Acid 2 Volt 2.5 Ah "D" cells for a total of 18 volts @ 2.5 Ah or 45 Watt hours.
The battery is made up of three strings of 3 cells.
Battery Connector is wired:

Pin	Wire Color	V @ Exp Owr	Function
1	Black	gnd	gnd
3	White	0.9	+Vsense
5	Red	6.8	+Vbat

The cells were made in 1983, about 24 years ago and are now dead beyond recovery. This is probably why these units are on the surplus market. Although it's relatively easy to rebuild the battery pack.
If +Vsense is connected to +Vbat then BAT STATUS lights when the button is pressed. But if the battery pack is removed then pressing BAT TEST does not light BAT STATUS.

Power Up

The HP E3617A bench supply can only put out 1 Amp which is not enough to start this Rb source. Using the B&K Precision 1786 that's good for 3 Amps worked fine. It came up at 27.7 Volts so that's what I'm using. Just after power up the following LEDs are on:
Set CLock = Red
Cold Osc = Red
External Power = Green.
starting current about 2.2 Amps.
In 5 minutes the Cold Osc LED turned off. (I think this the fast warmup option on the M100)
At any time pressing the Battery Test button causes it's green LED to turn on.

Feeding the 10 MHz out to the SR 620 counter in frequency mode gives the same display to 10 digits as the FTS4060 Cesium standard. The SR 620 is using a GPS disiplined SR PRS10 Rb standard for it's time base.

28 Oct 2007 after replacing the battery pack the supply current after the oven has stabilzed is just under 1.2 Amps. So if the main vehicle power failed after the O-1814 was warmed up it should keep running for over 2 hours. Charlie's comment above about "15 minutes" indidcates that the backup batteries have been damaged.

Still need to figure out the connector wiring for Time Of Day input and output.

Theory of Operation

From the POWER connector 16 to 32 VDC goes through a filter, the RESET breaker A5P1/A5J1 to the A2 Power Supply PCB that's below the battery and Rb standard.
From A2 to PS1 which is a DC to DC converter that outputs 28.0 Volt which is filtered by A5L1, A5D1 and A5C1 then goes to PS2 which provides the working voltages:
+28, +12, -12.3, +6.2 and +6.3 VDC.
The M100 outputs 10 Mhz to E2 of the active splitter that's behind the 10 MHz Out BNC connector. E3 is probably 10 MHz that goes to the A2 PCB which divides it by 2 generating a 5 MHz local reference which is compared to the 5 Mhz input. If the 5 MHz input is present that's fed to the Time Of Day PCB, and if not present the local 5 MHz is used for TOD.

Outputs

The outputs are:
Time Of Day
5 MHz
There are a number of 800 kHz signals and it's not clear if these have anything to do with the RFO or if they are just passed between the UHF-AM radio and the SDU.

Time Of Day

The A5A2 Time Of Day circuitry consists of two PCBs a mother board that has two connectors and a daughter board with no connectors to the rest of the O-1814. There's only one filter box behind R/T-TS, SDU and RT-1319 and there are only 5 terminals and E1 (ground). They are: E2, E3, E4, FL2 and FL1. Which may include: DC power, TOD data in, TOD data out, TOD clock in, TOD clock out, and some type of data valid indicator.

Time Of Day is in a standardized military format that's supported by many military GPS receivers like the PLGR and Trimpack units. The frequency hopping radios, like the RT-1319 (Have QUICK), SINCGARS (RT-1439, RT-1523) and others need Time Of Day to synchronize.
The KY-57 voice crypto does not need TOD, not does the internal crypto in the newer SINCGARS radios.

There's a good chance that in addition to the clock and data lines there is also a 1 Pulse per second output.

The Rx TOD and Tx TOD buttons are wired to the A5A2 TOD board. The Set Clock LED has one side wired to the TOD board and the other to the A2 board which controls the brightness of all the LEDs by means of the Bright, Dim, Off switch. The Rb interface is to the A2 board and it controls the Cold Oscillator, Fault, Ext Pwr, Bat Status, Bat Test front panel items.

A1-A1 Box behind RT-1319, R/T-TS & SDU

Feed thru	wire clr	VDC	4k7 Pull Up	4k7 Pull Dn	Description	RT-1319 Pin	R/T-TS Pin	SDU Pin	1 Jan 2009 TO 31R2-2GRC206-2-1
E1	Blk	0.000	na	na	chassis ground	11, 26, 27	E	26, 27
E2	Blk	0.000	na	na	chassis ground	11, 26, 27	E	26, 27	800 kHz Return
E3	Wht	3.121		2.7	1 MHz TTL out (modulation?)	31	D	nc	800 kHz Out
E4	Blu	5.39		4.3	? nc inside box	nc	nc	nc
FL1	Brn	+8.6		5.0	?	nc	J	nc	TOD Continuous
FL2	Wht/Red	2.4 0.3?³	5	0.0	Input (TOD?)	25	F	nc	TOD Out
FL3	Org	0.03		0.03	send TOD^₄	10 ¹	B	nc	TOD In
FL4	Yel	5.67		5.63	?	21	A	nc	+5 Out
FL5	Grn	5.39		4.42	?	32	H	nc	External 800 kHz Enabl
FL6	Wht/Grn	2.4 0.03?³		0.03	10 us pulse @ 32Hz (31.25 ms P)	nc	K	nc
FL7	Wht/Brn	0	0.2	0	?	nc	C	nc	nc

Note 1 - The PLGR Have Quick TOD cable uses pins 10 and 11 to load the TOD into the RT-1319/PRC-113.

Note 2 - It looks like the Signal Distribution Unit (SDU) connector has no signals to or from the O-1814. The RT-1319/PRC-113 does need TOD data and there are 5 lines going to it. The R/T-TS has ALL the signals. This box sits beteen the RT-1319/PRC-113 and the Signal Distrubution Unit.

Note 3 - These values are different than what was measured before the standby battery was replaced.
Note 4 - This is the TOD message, but is all ones. i.e. a positive TTL pulse 300 us long followed by a 300 us 0 volts.

Oct 31 2007 -the signal on E3 has two different pulse hi and pulse low times so looks like some kind of modulation. This is always on, not dependent on any button presses. The frequency is between 714 kHz and 1.5 MHz. Capturing some pulses in single shot mode shows :

Time Hi us	0.8		0.6		0.88		0.78
Time Lo us		0.6		0.5		0.4

After doing the Pull Down measurements the voltages on FL2 and FL6 are considerable different than they were when first measured. This may be related to having a working backup battery now and they a dead backup battery? The HP 54501 scope shows a Have Quick data pattern of FL3 that lasts about 277 ms when the SEND TOD button is pressed. It's a square wave, i.e. there's is no TOD in the O-1814 so it's sending a null string. But at least the function of one line is now known.

The pulse on FL6 is the same as before a 9 us wide pulse with a rep rate of 32 Hz. Don't know what that's about.

FL1 is interesting in that it's open circuit voltage is +8.6 (above TTL) and it comes down to 5.0 V when a 4k7 resistor is pulling it toward ground.

The RT-1319/PRC-113 has the ability to output it's TOD so it can load the O-1814 with the TOD. Military GPS receivers have two functions, the most commonly known are the functions relating to position, but the other is to load TOD into equipment that needs it.

Have Quick

Have Quick is the name associated with frequency hopping radios. It's usually applied to the UHF air band radios like the PRC-113 or VRC-83 that use the RT-1319 or to the aircraft radios. The details are not in the public domain and so need to be figured out. It seems that HQ equipment can both receive and send the TOD. There are different flavors of Time Of Day (TOD). But all of them have the same starting data so an older radio that's expecting the basic message will correctly load the front part of an extended message and ignore the remaining bits. This is know as backward compatible.

The SINCGARS time (TM 11-5820-890-3) is different from the Have Quick time. In SINCGARS the date is just 2 digits which are the last two digits of the annual Julian day number. There are two calendars, one for regular years and one for leap years. For both Feb 28 is day 59. For a regular year April 10 is day 00 as is July 19 and October 27. In a leap year the 00 dates are April 9, July 18 and October 26. The SINCGARS radios resynchronize their TOD for the current channel every time the Net Control Station transmits. The Have Quick radios do not do this. The crystal oscillator in the Have Quick radios is only good for about 4 hours then has drifted to where they no longer will work. Have Quick radios can get an off the air TOD but it's a manual procedure.

The Wiki Have Quick page says the Radio needs:

accurate Time Of Day (TOD)
Word Of the Day (WOD)
NET number

All of the above three elements are needed for the frequency hopping.

Basic Have Quick TOD Message (STANAG 4246 HQ1)

Sends the Hour, Minute and Second. Message is less than 1 second long and uses BCD coding. Very much like the WWV and WWVB time codes that take one minute to send. Maybe if it's 3:46:27 pm (15:46:27 UTC) the BCD values sent would be 1, 5, 4, 6, 2, 7 and each digit would take 4 data bits for a total of 24 bits. But if self clocking Modulation is used then you can think of that as using 2 message bits for each data bit and so the number of message bits is doubled to 48. For example a 1 data bit is a message of "10" and a zero data bit is a message of "01". This scheme is self clocking so a seperate clock wire is not needed. That's consistent with the PLGR to Have Quick cable having only signal and ground wires. The crypto fill system uses seperate clock and data lines.

Have Quick II (STANAG 4372)

The basic HQ message then Day of the Year as 3 BCD digits (3 * 4 * 2 = 24 bits) and two digit Year (2 * 4 * 2 = 16 bits)

GPS Based Have Quick (ICD-GPS-060) aka PTTI HQ

The HQII message then Time Figure Of Merit. The TFOM covers the range of 1 ns to 10 ms. On the PLGR GPS receiver it's expressed as a single digit from 1 (<= 1 ns) to 9 (>= 10 ms). So 1 digit * 4 bits/BCD digit * 2 bits message coding = 8 bits.

The ICD-GPS-060 spec has two protocols.
One is a two wire differential (maybe like RS-422 or RS-485 haven't checked that) and the other is a single ended TTL level Extended Have Quick.
There's also a timing fault wire that has a TTL level for OK/NG.

Description	1PPS	Data	duration ms	End since 1PPS ms
Data Frame	10V 20 us	400 "1" char	240	240
"	0V	16 bit Sync word	9.6	249.6
"	0V	2 digit hour	9.6	259.2
"	0V	2 digit minute	9.6	268.8
"	0V	2 digit second	9.6	278.4
"	0V	3 digit DOY	14.4	292.8
"	0V	2 digit Year	9.6	302.4
"	0V	1 digit TFOM	4.8	307.2
"	0V	low	692.8	1000

16 bit Sync word		11E9 hex

"1" char		300 us hi
		300 us lo

digits		1667 bits per sec 600 us per bit
digits		4 parity bits then 4 data bits

The parity bits come from a table that's based on the idea of not only detecting an error but of correcting as many bits of error as possible.

Extended Have Quick (STANAG 4430)

The GPS HQ message then bits for a more accurate TFOM, Continuation data, Leap Seconds and an End Of Message terminator. Maybe if there is no message terminator the message can be repeated with the continuation data different in each sussive data packet?

HQ Physical Layer

Depending on the flavor of HQ it may be as simple as a ground and a data wire. The O-1814 has two front panel buttons RCV TOD and SEND TOD. The SEND TOD might be the simple method. The RCV TOD could also just enable one frame of reception of a data stream but it's more likley that there's a more complex physical layer like a seperate wire for a "load me" request and maybe a seperate clock wire.

Backup Battery

The battery is a backup for the main DC input and will keep the Rb alive and the TOD will be maintained as long as the internal battery lasts. When the front panel RESET/OFF switch under the red cover is turned off the internal batteries do NOT power the unit and it forgets everything. When the switch is on and there is external power the battery is float charged. If the external power fails with the switch on the batteries are used to power the O-1814. They are rated at 2.5 AH and the unit draws about half of that so they should last at least a couple of hours.

These Cyclon circular "D" cell size lead acid batteries were developed for standby applications very similar to this where they power the equipment when the power fails and are recharged immediately after the power comes back on line. They will survive being drained to zero capacity in this application IF THEY ARE RECHARGED A SHORT TIME LATER. This is far different from other batteries which will be ruined if discharged to zero. But if the Cyclon is discharged to zero and left discharged for some time it takes a special procedure to attempt recovering their capacity.

If a user of the GRC-206 system shut it down from the Power Distribution Unit, i.e. turned off all the radios and the O-1814 from the PDU and walked away the backup battery would power the unit until it was flat. Turning off the GRC-206 system at the Power Distribution Unit does not turn off the RFO. But it would not get recharged until the next mission thus degrading the capacity after the next charge. But if as part of the power down sequence the O-1814 was turned off first then the batteries would drain at their shelf life rate. They could sit for many years if starting out with a full charge. 2009 - It's still a mystry about the poor performance of the RFO.

Jan 2009 - The manuals mention that as part of the receiving inspection the internal On-Off switch should be turned on. This switch is located in the backup battery circuit. It must be on for the backup batteries to work.

Description

The OFF/RESET switch is a DPDT type that completly disconnects the external power from the O-1814. But it also has a ground connection on the input side that gets opened when the switch is opened. This gives a signal to the A2 PCB that turns off the internal backup battery (it may be as simple as the backup battery ground connection is through this wire).

There's a sticker on the right side that says:

WARNING
REMOVE BATTERIES BEFORE
SHIPPING OR INACTIVE STORAGE
OF 30 DAYS OR MORE

It seems that someone didn't follow that advice. But the batteries have not yet started to leak.

The battery box is lined with the same Rodgers Poron Urethane Foam that's used for skin diving suits. Although there is a vent hole that's aligned with a plug in the bottom cover it appears to be a safety vent. Thre's no provision to keep acid fumes from getting to the electronics, but it may be that the Cyclon cells are well sealed.

There are more patents on the Cylindrical Lead Acid cell than are shown on my Battery Patents web page.

26 Oct 2007 - received replacement backup batteries. The dead battery packs were marked 810-0113CC for each 3 cell pack, but the new ones are the 0810-0103 and appear to be the an exact replacement. Note these are the cylindrical cell CYCLONs the dead ones were made by Gates and the new ones are branded EnerSys.

As received the 3 cell packs were at 6.15 Volts Open Circuit (VOC). The applications manual says that's about 60% capacity. 1.95 V/cell is 10% and 2.14 V/cell is 100%.

The recommended float charge is 2.30/2.25 Volts/cell or 20.7/20.25 Volts per pack of 9 cells. With a bench P.S. set to 20.5 volts the 9 cell pack is drawing about half an amp in Constant Voltage mode. After 45 minutes the float current is down to 318 ma. The next morning is was down to about 20 ma. After powering from the external supply and waiting for the Cold Oscillator light to go out the external supply was disconnected. The Ext Power lamp stayed green for an hour and the battery terminal voltage was 16 Volts or below the zero capacity value so the test was stopped and external power restored. This should charge the battery.

29 Oct 2007 - after more than 24 hours of charging the battery voltage is 19.9 measured with the main power switch off so there is no float charging. The battery pack voltage varies from 17.55 at zero capacity to 19.26 for 100 % capacity, so 19.9 Volts is a little more than 100 %. It may be that a small load or some amount of time needs to pass after taking the batttery off float charge is needed to get an accurate reading of capacity. In any case the float charger has filled up the battery.

O-1814/GRC-206 Reference Frequency Oscillator

Rubidium Frequency & Time Standard

Background