Re: HP 8566A large amplitude error in 2-22GHz range
ok the fact that peaking has worked and reduced the loss substantially is a good sign. Peaking works on the FM coil of the YTF
now based on your description the LO seems to be unlock all the time and it is wobbly. Therefore, I strongly recommend
you focus on fixing the LO loop and forget the amplitude loss for the moment because at this point I believe it can be due to the lack of stable?
Also note that if the YTO has low amplitude which can cause unlock LO, it can also cause loss of amplitude because the mixer's conversion loss will increase quite a bit
if the LO has low amplitude.?
so focus on the LO loop for now. I have not worked on any of these 8566/8568 machines but I am sure the structure is universal.
The causes of LO unlock are "usually" one of two things (1) the YTO itself is bad for example it has very low amplitude at some frequencies
(2) the reference section (there are PLLs and multipliers in the reference loop) is bad so the reference frequency that YTO is supposed to lock to is bad
if you have another SA, you can check out the LO amplitude output directly and see if it has any "holes" in it over the full band of YTO
if it does not (flat strong amplitude over the whole band of YTO) then you are good
I am not familiar with the reference section of this analyzer but I am sure there are points that you can connect another analyzer and see if the frequency is rock solid stable or not. One basic test is to connect the Calibration signal of your analyzer to another analyzer and see if it is stable and solid or it is wobbly.
it is quite possible that machine is very old and needs some recapping in the reference section and YTO drivers. that should be easy to do
after the LO is fixed we can go back and see how the amplitude is.?
?
|
Re: Weather resistant, covered Litz wire wanted.
Hi, Lize wire is to compensate for skin effect, and the higher the frequency the shallower the "skin" so more wires equals more skin serface and less losses.
Regarding a loop aerial/antenna for reception, coating dsc or dcc litz wire with transformer varnish sounds reasonable.
You will need to protect it from sunlight, some kind of outer tube round it then flood the whole lot with varnish might work but would need a rigid frame to prevent twisting and cracking the varnish.
best of luck , 73s,
OldBlueBear-G4JHT-Dave
|
Re: [test-equip-trader] 8671, 8672, 8673 support kit
Bonjour all: Pete is under severe stress, perhaps due to health problems.
I kindly suggest the group to take this into consideration and cut Pete some slac.
I humbly suggest that we end this thread.
Your friend and fellow HP/TEK aficionado,
Bon courage et bon Sante,
Jon
|
Hello Folks,
I just spent 10 minutes of my limited time digging through whiney messages. So, f### it, Ill invest another ten minutes to point the esteemed quorum at a general fact which, we TMI heads, might not like.
A lot of old TMI is worth less than the sum of its parts. My ex lab gypsy recently experienced this upon trying to import a DANAHER 575 curve tracer to Macao. It was a honest customs official who did not try to pilfer the unit, but had her over for a chat. He at first believed that she had found the mother of all import scams, and was utterly shocked AND very unhappy when she showed him that the 575 will actually see use as it stands.
Same for Nixie tubes. I have, in my lab, two Systron Donner Microwave counters (if anyone sees Artek, nudge him from me - dude, I REALLY need me manual back. Please, please man!!! Its now three years!!!! Ill pay shipping. I pay processing. PLEASE; give it back as I did not make a copy before sending it to you).
If we tear them down, the Nixies are worth more than the whole unit. Actually, I have a design called Nixie-be-gone floating around my head, just never came to it...as nixies are maintenance hogs anyways, why not let customers release the Nixie equity to fools who desire it, and improve TMI dispatch reliability.
Either way, with gold prices at a record high thanks to a dude named after a ceramic lavatory equipment, we should expect a lot more scrapping soon. Sorry to bring the depressed news, but 500USD is less than 10 grams of gold.
Tam
- - - - -
With best regards
Tam HANNA
Enjoy electronics? Join 13500 followers by visiting the Crazy Electronics Lab at
toggle quoted message
Show quoted text
On 07.12.20 04:20, Dave McGuire wrote: On 12/6/20 9:32 PM, Dave Miller wrote:
I am good on a lending library. ? It's a good idea.
I don¡¯t agree with your approach with Pete. ? That is your prerogative.
He has helped many of us out on our quest for nice test equipment. ? I myself have purchased about $10,000 worth of stuff, possibly a bit more, from Pete.? Most of those purchases were good ones.? Not all, but most.? But, importantly, that's not what we're discussing here.
I doubt he got all his stuff for free. ? No one knows how he got his stuff.? But he has claimed to me several times that at much of it did, actually, come for free.
Expecting him to give stuff away for free. ? No one EVER said anything about him giving it away for free.
The value of a item is what the sell and buyer agree to. ? Correct.
Not all the time they agree. ? Correct.? But the line that was crossed was the public threat of destroying the item if his (very high, in this case) price demand was not met.? As I stated earlier, it was inappropriate, uncalled for, and transparent.? I stand by that assertion.
? This is not the first time this has happened, and it's not the first time he's been called out on it.? In fact, it's not even the second, or third time.
I find this a great resource and with my growing collection of HP gear I hope to continue being allowed to read and post but am disappointed with some of the posts today. ? As am I.? Different ones, apparently.
??????????? -Dave
|
Re: HP 8566A large amplitude error in 2-22GHz range
That's at 18GHz I believe.
|
Re: HP 8566A large amplitude error in 2-22GHz range
The cable is 5ft long, so there is some loss. I want to say it's 1.1dB loss per foot, although I could be slightly off. I can find out the exact number if that's important
|
Re: HP 8566A large amplitude error in 2-22GHz range
amirb, thanks for the info on the likely causes! I need to spend some time working through the RF operation, but this should focus my search. The frequencies are very close. Not precisely correct, but they don't look far off. They're pretty stable, but they do bounce around a bit. This is especially noticeable when I set the frequency span low.
Don, the noise floor is exactly as you describe. When I perform a preselector peak I get:
2.2GHz: -1.7dB
4: -1.7
5.7: -3
5.9: -2.8
8: -3
12.4: -5.5
12.6: -14
18.5: -13.5
18.7: -13.5
20: -11.9
That's as high as I can go. I'm not sure if this is normal, but the amplitude moves around a lot during "peaking!". The cable is of high quality, but is only rated to 18GHz.
|
Hi Dave,
I bought a Masters Communications RLB1500 from Amtronix, to use
with my HP8935 you can see it here...
cheers Tim
On 7/12/20 2:34 pm, Dave Ireland wrote:
toggle quoted message
Show quoted text
Is anyone on this group have a return loss
bridge for sale or know where a unit with HP type specs can be
obtained at a reasonable price?
I want to use it for sat-com stuff so the
top end frequency does not have to be above 500MHz but it
would be nice to match it to my spec an (Agilent E4401B) up to
1.5GHz.
Thanks
Dave
|
Re: 9122 and 9122C - HPIB Connected DiskDrives - Scrap or save?
Robert, just for clarification the HP 9122 is a dual 3.5in. floppy drive (360KB per drive) with a GPIB connector. The HP9122C is the same but both floppy drives are 1.2MB.
They were used a lot with the HP ETE and technical workstations in the 1970¡¯s and 80¡¯s. They can be used to load software and sisters onto the older HP ETE.
Don Bitters
|
Re: HP 8566A large amplitude error in 2-22GHz range
1) What does the noise floor of the 8566A look like with no signal connected after IP (Instrument Preset).
hint: the top of the noise floor from 2-5.8GHz should be flat and about 65dB down from the top graticule, from 5.8-12.5GHz almost flat and about 60dB down from the top graticule, from 12.5-18.6GHz - tilted up about 4dB and about 54dB down from the top graticule, from 18.6-22GHz - tilted up about 6dB and about 48dB down from top graticule. The gain is stepped up about 6dB for each band from 2-22GHz, which causes the noise floor steps.
2) What happens when you put an external signal from your 8340Bon each of the bands I listed above and then do a PSP (preselector peak - a front panel key)?
Better yet put a signal in at the low, mid, and high ends of each band, but not right at the band edge - 100-200MHz above the lower, and below the upper edges will work just fine.
3) What kind of cable are you using to hookup the 8340 and 8566? Is it rated for 22GHz?
Don Bitters
|
I bought my perfectly in Cal hp 86205a for $200 so...
|
From Kevin Custer of Repeater Builder fame:
|
Thanks
I am never sure of the "real"
performance when buying off ebay but if you have had success with
them I will go and see if I can hunt them down.
Dave
On 7/12/2020 11:58 am, amirb wrote:
toggle quoted message
Show quoted text
I have several of those cheap ones on ebay and the good ones work
perfectly fine up to 1GHz or even more honestly.
there is one made by a Russian seller which I find it the best. it
also comes with metal casing and looks very professional
claims 3GHz but it's really good to 1-1.5GHz unless you can accept
lower directivities when yeah 3GHz is not horribly bad?
also there is? mini circuit zfdc-20-4 (cheap on ebay) that works
fine up to 1GHz
i think it has a 75 ohm version too
i do have a HP 86205A up to 6GHz which is dear to me ;-)?
|
I have several of those cheap ones on ebay and the good ones work perfectly fine up to 1GHz or even more honestly.
there is one made by a Russian seller which I find it the best. it also comes with metal casing and looks very professional
claims 3GHz but it's really good to 1-1.5GHz unless you can accept lower directivities when yeah 3GHz is not horribly bad?
also there is? mini circuit zfdc-20-4 (cheap on ebay) that works fine up to 1GHz
i think it has a 75 ohm version too
i do have a HP 86205A up to 6GHz which is dear to me ;-)?
|
Is anyone on this group have a return loss
bridge for sale or know where a unit with HP type specs can be
obtained at a reasonable price?
I want to use it for sat-com stuff so the top
end frequency does not have to be above 500MHz but it would be
nice to match it to my spec an (Agilent E4401B) up to 1.5GHz.
Thanks
Dave
|
Re: [test-equip-trader] 8671, 8672, 8673 support kit
On 12/6/20 9:32 PM, Dave Miller wrote: I am good on a lending library. It's a good idea. I don¡¯t agree with your approach with Pete. That is your prerogative. He has helped many of us out on our quest for nice test equipment. I myself have purchased about $10,000 worth of stuff, possibly a bit more, from Pete. Most of those purchases were good ones. Not all, but most. But, importantly, that's not what we're discussing here. I doubt he got all his stuff for free. No one knows how he got his stuff. But he has claimed to me several times that at much of it did, actually, come for free. Expecting him to give stuff away for free. No one EVER said anything about him giving it away for free. The value of a item is what the sell and buyer agree to. Correct. Not all the time they agree. Correct. But the line that was crossed was the public threat of destroying the item if his (very high, in this case) price demand was not met. As I stated earlier, it was inappropriate, uncalled for, and transparent. I stand by that assertion. This is not the first time this has happened, and it's not the first time he's been called out on it. In fact, it's not even the second, or third time. I find this a great resource and with my growing collection of HP gear I hope to continue being allowed to read and post but am disappointed with some of the posts today. As am I. Different ones, apparently. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: HP 8566A large amplitude error in 2-22GHz range
I am not familiar with 8566 but I have repaired quite a few other high frequency SAs by HP
I think most probably the structure is the same for high bands, meaning a pre-selector YIG filter and then a harmonic mixer
so here are my thoughts and suggestions/questions
1-Although the intermittent YTO unlock issue must be addressed and fixed before anything else, but for a moment we can ignore it
if you are reading correct and stable frequencies when you are doing those amplitude measurements. Are you?
if it is not stable and is wobbly, then troubleshooting amplitude problem is useless, the YTO lock must be fixed
but if you happen to have lock at those measurements, then based on my experience your errors pretty much
most likely due to the YTF (yig tuned filter) in the high band and a little less likely but still possible in the harmonic mixer or the high band switch
these three components are the main and usually only pieces causing the problem in high band. I dont know if they are integrated together
in 8566 like in 8563 for example or they are separate. If they are separate, it is easy to follow the RF path and find out where the
signal is losing its amplitude. In most cases that I have dealt with the YTF is the min culprit and then the high band switch.
it could the YTF driver circuit is bad and is not tuning the filter to follow the signal or it could be the YTF itself which would be unfortunate.
Is there a "amplitude peaking"? or pre-tune
or pre-selector peaking .... option in the menus? usually high frequency SAs have a peaking button in menu that tries to maximize (fine tune the YTF)
at the center freq. you can try that, it may be called various names but you will figure it out. But if the filter is way out of tune that?
option cannot bring it to center.?
the only way to really figure out the issue is to open it and follow the signal path with another SA or at least a power meter
|
Someone was looking for a power cord that doesn't plug in, for a repair. I just bought a batch on Ebay, and the seller has one lot left.
which is a lot of 25, 18AWG, eight foot cords for $46.66 including shipping. That's less than $2 each, and they aren't as easy to find as they once were. I figure that by the time I replace all of the worn or damaged cords on my stuff, I might only have a couple left. I just wanted to let the group know there was one lot left, from and old line surplus dealer..
|
8594L is a totally different instrument
|
just to make sure, did you connect the comb output to the rf input for the calibration?
what happens if you connect t he comb output to input and try to just look at it??
|
OldBlueBear-G4JHT-Dave
Froggie the Gremlin
Dave McGuire
Michael A. Terrell