I'm bringing on-line a new used system as above. Hoping some of you folks may be well-versed on the topic. I have the setup as above, still just doing qualification runs. Using an HP 83623B (20 GHz) generator as the RF source. It seems to be working well in that regard. I have the original sample file of the TRANSFER STANDARD with a current due date, and ID #, and using the sample cal factors.
And using HP 436A as power meter, and testing an HP 8481A as the UUT. I have been walking through cal of the 8481A with above setup. I am occasionally getting good, believable cal factor readings, but they are intermittent. At first I was thinking my TRANSFER.DAT wasn't good, until I started getting varying cal factor results, some good, some completely out to lunch. Some readings, the cal factor is blank, after numerous errors during the cal. Some of the erroneous cal factor readings might be 150%, 300%, 400%, others 0%. The observed power readings on the HP 436A are all good (typically between 0.900 mW and 1.05 mW). The DC null at the beginning of the cal is good and stable, and the coarse and fine null adjustments seem to be able to normally adjust the zero. And as I watch the null meter during the automated cal, It seems to stay centered at about zero. Everything on the front panel of all instruments looks like a good reading. But the cal factor results look bad. The one other thing is that the procedure calls for an 1805B, which I have. But the available driver is for an 1805A. I wonder if there is enough syntax difference that an 1805A driver doesn't work properly with an 1805B.
Any knowledgable thoughts appreciated.
I've used a lot of Tegam mounts, and cal a lot of them as well.
In my 3+ years working with the assorted variety of Tegam stuffs, the ONLY thing I have EVER used my 1805B for was a warmer for my other mounts. If you can swing an 1806 or an 1804, I'd get one of them instead.
But, if you have the support from SureCal, give them a call. They are always very helpful and can at least point you in the right direction regarding the drivers, etc.
Good luck!
I never used Surecal to do mounts, but have worked a bunch with the 1805s. The A and the B are the same so I don't think that's your problem. I think from reading you said you had the cal factor file for the standard mount, it isnt missing any data points that might be causing the problem. Other than that I don't have much for you, sorry.
I've had this issue before and checked the cables on a spec an and found out I had a bad one. If that doesn't work I usually calibrate the transfer standard again to get updated factors since I never know if it was turned off without my knowledge.
Let me ask a question.... And this might help me. Are there cal factors stored in the 1805B other than those that are supposedly downloaded from Surecal when you do the cal. Because it looks like what I have is all of the readings (in an analogue sense) are working. That is, I see good numbers of close to 1 mW on the 436A (with obvious downward drift as you head up toward 18 GHz (down to around 0.75 to 0.8 mW). And I see a good null on the 1805B. It stays at 0 null point. But the cal factors generated for the UUT are all over the place. So I've been wondering (after about 30 experimentation runs yesterday) if there are some missing cal factors for the transfer standard that are resident already in the 1805B. Or perhaps I have bad memory locations on the 1805B. It acts as if the readings are working good, but the math relative to cal factors is not good. If the cal factors are downloaded each time I freshly run the Surecal procedure, then possibly there is a bad memory chip that maybe needs to be replaced. If there is already some resident memory numbers, then maybe I have corrupted memory.
Also, to PSIRUS716 - Can you elaborate about your second sentence. I just got this system used, and trying to functionally verify it. I don't have real cal factor numbers. I am using the generic set that comes with Surecal's template for TRANSFER.DAT file. I updated the due date in it, and verified that the numbers are reasonable "fakes". I even tried shifting the 10 MHz down as low as 65% CF. The number that are odd to me are that at (for example) the generated CF for the UUT (HP 8481A that is in good cal condition) is 0%. some of the other CF's go up to 230% or so. It is all over the map, as if (even though I am getting good nulls, and 436A readings) there is something corrupted in the computations. I spent considerable time experimenting on the TRANSFER.DAT file (stores the CF for the 1807A). But it is acting strangely., I'm hoping to get a good clue from one of you folks experienced on this system.
One more note. I have a bunch of other work to do as well (I have to actually get some calibrations done as opposed to working full time on our new capability - once it is running). So I sent an email to the OEM yesterday, and I am going to send the entire system in for a check. I'll get minimal checks done on everything except for the M1110 (the primary standard). I'll get an accredited cal for that, and use it to cal the system. For you experienced users out there - does that sound like the legitimate path?
There are no cal factors stored in the 1805. The 1805 is the bridge used to balance the rf level of the generator. if the bridge is set to 1 mw you should have 1 mw at the output of the splitter beween your uut and primary standard. from that point on it is all math, the standard cal factor and the reading on the uut power meter. It really sounds to me like you don't have a complete table of standard cal factors, i.e. you are testing frequencies on the uut that don't have a corresponding entry in the cal factor table.
Ok, since you are going to stick to the 1805B.....
I would not waste the $$$$ sending it all in to OEM for cal/repair. It really sounds like you have an issue with the 1805B. Try pulling the manual from the Tegam website and running through the cal for the 1805B. It is a pain to do (I've done it a few times). Make sure you check all of the capacitors in the power supply, they are notorious for going out, as they are in all the RF bridges.
Even using the "default" values in the transfer.dat (even if you are missing cal factors for some of the frequencies, SureCal will interpolate), you should still get somewhat believable numbers, unless of course, you have an issue with the 1807A itself, which is possible. That's the downside of using the 1805, you don't get to see the multimeter readings like you would using the 1806/04 to see if they are around where they need to be.
Ouch, an accredited cal from the OEM, that's a huge chunk of change, but unavoidable unless you can find another source.
Good luck.
First and foremost, the M1110 terminating mount is the highest level mount you have and should be calibrated by NIST for the best uncertainties (<1%). If you don't mind uncertainties from 1-2% send it to Tegam. The M1110 will then be used to calibrate your feed-through mounts like an F1116 or 1807A. The feed-through mounts will then be used to calibrate 8481A, 82A...etc.
If you are using an 1805B, ensure the signal from the 3325B is connected to the proper AM connector on the 1805B rear panel. I believe it is the "AM-not" connector. The profile for feed-through type mounts in Surecal requires that Power Monitor 1 be the UUT, which in this case will be the 1805B, Power Monitor 2 be the Standard, the 3458A.