One of the main reasons for getting the TAC2 was to use the 1PPS output in order to discipline a local oscillator. This will create a highly accurate frequency standard. Brooks Shera documented this in an article for QST magazine and this is where I got the idea.
In the discussion that follows, I will document my attempt to create this project.
Motorola Time standard
While I was preparing to start with the clock discipline standard, I found a nice item on eBay. It was listed as a Motorola Time Standard. I immediately jumped at the auction and I got the unit for $210.00.
Note: For the discussion that follows I will refer to this unit as the "MTS".
This is a small description of the MTS:
It is housed in a 2U 19' Standard rack mount case.
In the front it has (from left to right):
In the back of the unit it has (from left to right):
Inside the MTS I could see a GPS receiver of some sort and a TCXO in a single PC board.
In order to power the unit I got a Acopian 48V 5A Power Supply from Ebay. This is a very stable power supply which includes two front panel meters for Voltage and Current display. The unit came pre-calibrated for exactly 48.0 volts.
A I connected the MTS DB25 output to my Windows machine that was running TAC32. Then I connected the MTS "Chassis Ground", "Battery Ground" and the "+(positive)" power supply terminal to ground. Then, I connected the negative power supply terminal to the "-48V" terminal on the MTS.
The TAC32 software received data from the MTS right away. It turns out that I had a Motorola VP 6 Channel GPS receiver.
So, now I had two GPS receivers and the MTS was already giving me an output of 5Mhz!
When I asked the people at the TAC2 mailing list about the MTS they suggested that it was probably from a cellular base station. This was based on the power supply requirements.
In order to feed the TAC2 and the MTS, I purchased from eBay a "HP 58535A" Distribution Amplifier. Although the people at the TAC2 list suggested that I might build a cheap distribution amplifier from Radio Shack parts, I opted to buy the HP unit. I didn't want to take a chance in blowing up by antenna or my TAC2!
The cost of the HP58535A was $100.00 which was expensive but when I received the unit I was not disappointed. It was very well built. I suggest, if you need a distribution amplifier, you get the HP unit. It's worth it.
Actually I bought a second HP 8 port distribution amplifier from Ebay and I am using that one now. It is a 58517A. The 8 port unit allows for the outputs to have from 4-30 volt DC bias. None of this bias is passed on to the output since the unit is powered from a separate DC SMC connector. This allows for almost any type of GPS receiver to be connected and to use a single antenna (in my case a Motorola antenna).
The one problem with the MTS was that the 1PPS signal was coming in on the RTS pin and not the DCD. A RS232 breakout box from Radio Shack fixed that.
The Brooks Shera QST article described the discipline of the HP 5328A counter. I thought this was a good place to start so I got three units from eBay. One is only good for parts but I have managed to get the other two online. These units are very well made and I can see why Brooks picked this counter for his project.
I also picked up a HP5370A Frequency counter. If you do high accuracy frequency measurements I strongly suggest that you pick up one of these (or a newer HP5370B). This is an amazing unit. It has 12 digit resolution and accepts either 5Mhz or 10Mhz standard frequency inputs. It also has an internal OCXO but you probably don't want to use that.
I am feeding the 5370A from the Motorola 5Mhz GPS standard and it likes that very much. It also doubles the frequency and gives me a 10Mhz output which is great! Now I have a 5Mhz and a 10 Mhz frequency standard. Cool eh?
HP Z3801A Frequency standard I got the HPZ3801A from eBay on July 15, 2002. There were some power problems but the unit is operational as of July 19. Here are some preliminary notes on the unit.
I would like to thank Mr. Tom Van Baak for providing a link to pictures of the power connection.
Also, Mr. Bill Jones, K8CU, for his straight forward description. Bill's web site has the definitive reference on this unit. You can find it here: www.realhamradio.com
Now, for the details:
The correct connection to the power supply is:
Power supply plus side(+) to the Red
wire which is normally the top part
of the Z3801A connector.
The connector on the unit can be easily turned around so the black and red wires could be anywhere. Always make sure which wire you are connecting to.
Now, for the power requirements. The unit, has a sticker in the back that points out two different power connections. The option that has a circle filled in (indicating that it is enabled) says:
Well, that is outright wrong! My unit required at least 52V to power up. 38-60VDC my foot!
Inside the unit there are a few test points on the PS board which can help with power supply problems.
If you are looking at the +15V test point, when the power is inadequate, the voltage will hover around 3-5 V. When the unit gets enough voltage, there will be a distinct jump to around 14.8. I just gave the unit 53V and the voltage inside is now 15.158 and holding steady.
So, after getting the unit properly connected, I downloaded the HP software from the net, and stared up the unit. It is now working perfectly.
Many thanks to all who helped with your comments and suggestions.
I decided that until I build the my own disciplined standard, I would get another HP Z3801A. It seems that my MTS and the Z3801A did not always agree after all. Sometimes I would see a discrepancy of as much as ±7mhz.
This might not seem like much but I was still wondering which machine was to blame. So, I emailed K4DLJ, send him the money, and I got another Z3801A.
The setup was the same as the first. Open, clean, convert to RS232 and close. Again, www.realhamradio.com was a great source of help.
I noticed, that www.realhamradio.com is also selling a software called GPSControl. I send the money (via Paypal) and I downloaded the software. It was a very simple procedure to set the software and the results are here.
Not surpisingly, the two HP Units agree completely and the MTS is again off. I guess I know who is wrong here :)
After getting all this equipment, it is getting time to start building. I created a block diagram of the unit I am planning build so I can have a reference point.
You will notice that I am planning to include a Single Board Computer (SBC) that will be running Linux. This box will be a stratum 1 NTP server.
Some other features are: Front panel LCD display and keypad access.