Re: 8566 to measure 100W FM transmitter
Thanks, if the transmitter has a RF sample I will use that
|
Re: 8566 to measure 100W FM transmitter
I might as well add my two cents.? I use a 30dB attenuator (300W) which feeds a 10dB directional coupler the tap goes to the SA and the direct goes to a HP 438. ? As many have mentioned be careful of the signal to the SA.? But if you are wanting to accurately observe the transmitter a direct connection is the only way to go, no pigtails. ?Any other method could impact measurements. ? Steve ?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of Bob Albert via Groups.Io Sent: Thursday, May 23, 2019 2:41 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] 8566 to measure 100W FM transmitter ? There are several safe ways to do this.? What I do is put the transmitter on a dummy load and couple to the analyzer via a coaxial cable with clips on the other end, shorted together.? That makes a pickup loop which I couple to the dummy load, which usually has enough leakage to do the job. This method is frequency dependent but for a transmitter the bandwidth is so small so as not to matter.? Start with the loop a considerable distance from the dummy load and gradually bring it closer until you get a satisfactory signal. If you want a more stable, repeatable signal, use instead a direct connection via a simple attenuator, probably consisting of a 10k resistor in series with a 50 Ohm resistor.? That will reduce the 70V from the transmitter to less than one Volt.? You can do this with three connectors.? The resistors can be hanging in space.? The insertion loss of this network will be much less than 1 dB.? The 10k resistor will dissipate a maximum of one half watt so use at least a1 Watt rating.? Or raise it to 50k for a dissipation of 100 milliwatts and even less loss. On Thursday, May 23, 2019, 2:10:33 PM PDT, tmillermdems <tmiller11147@...> wrote: I would use a directional coupler such as a HP778D to get a 20 dB sample. Then pad the sample port with a 30 dB pad. That will get you to 1 mW into the SA. Note that the 778d is rated for 50 watts cw. There are couplers made for FM transmitters. You can also find a Bird sampler slug that will go in a Thruline wattmeter. Another option would be to build one out of coax. It is a pretty simple device. Maybe just a short stub antenna and grab it off the air? Just be careful not to overload the SA input. The mixer is not very forgiving. ? ? On 5/23/2019 3:50 PM, DW wrote: I should add the 100W will be driven to an antenna on a tower, I don't have a sampler so I'm limited to what I can sample, hopefully the transmitter has a sample port
|
Re: 8566 to measure 100W FM transmitter
Thanks for all the helpful inputs, I think I should probably invest in a portable analyzer that is not so heavy and a sampler.
What I will use with what is available is the JFW 80db step attenuator, a coaxial cable connected to it with a short piece of wire on the end going through a resistor to the coaxial shield, one wire will have a 400 ohm resistor and another 200 ohm. I might also bring along an oscilloscope to be on the safe side as it has selectable 50 ohm inputs. Based on what I gather so far with my setup I should be ok if you agree although it could be better. If anyone suspects a hazard with my setup which could damage the 8566 let me know. Thanks
|
Re: 8566 to measure 100W FM transmitter
There are several safe ways to do this.? What I do is put the transmitter on a dummy load and couple to the analyzer via a coaxial cable with clips on the other end, shorted together.? That makes a pickup loop which I couple to the dummy load, which usually has enough leakage to do the job.
This method is frequency dependent but for a transmitter the bandwidth is so small so as not to matter.? Start with the loop a considerable distance from the dummy load and gradually bring it closer until you get a satisfactory signal.
If you want a more stable, repeatable signal, use instead a direct connection via a simple attenuator, probably consisting of a 10k resistor in series with a 50 Ohm resistor.? That will reduce the 70V from the transmitter to less than one Volt.? You can do this with three connectors.? The resistors can be hanging in space.? The insertion loss of this network will be much less than 1 dB.? The 10k resistor will dissipate a maximum of one half watt so use at least a1 Watt rating.? Or raise it to 50k for a dissipation of 100 milliwatts and even less loss.
Bob
On Thursday, May 23, 2019, 2:10:33 PM PDT, tmillermdems <tmiller11147@...> wrote:
I would use a directional coupler such as a HP778D to get a 20 dB
sample. Then pad the sample port with a 30 dB pad. That will get
you to 1 mW into the SA.
Note that the 778d is rated for 50 watts cw.
There are couplers made for FM transmitters. You can also find a
Bird sampler slug that will go in a Thruline wattmeter.
Another option would be to build one out of coax. It is a pretty
simple device.
Maybe just a short stub antenna and grab it off the air?
Just be careful not to overload the SA input. The mixer is not
very forgiving.
On 5/23/2019 3:50 PM, DW wrote:
I should add the 100W will be driven to an antenna on a tower, I
don't have a sampler so I'm limited to what I can sample,
hopefully the transmitter has a sample port
|
Re: 8566 to measure 100W FM transmitter
The reason for needing to measure this FM 99.1 transmitter is complaints of spurs so all I will be checking is if the expected signal is there and no spurs of that signal by turning the transmitter on or off. Amplitude is not critical just need a safe operating range into the 8566. Yes this analyzer is extremely heavy but I don't own a portable analyzer.
|
Re: 8566 to measure 100W FM transmitter
I would use a directional coupler such as a HP778D to get a 20 dB
sample. Then pad the sample port with a 30 dB pad. That will get
you to 1 mW into the SA.
Note that the 778d is rated for 50 watts cw.
There are couplers made for FM transmitters. You can also find a
Bird sampler slug that will go in a Thruline wattmeter.
Another option would be to build one out of coax. It is a pretty
simple device.
Maybe just a short stub antenna and grab it off the air?
Just be careful not to overload the SA input. The mixer is not
very forgiving.
On 5/23/2019 3:50 PM, DW wrote:
I should add the 100W will be driven to an antenna on a tower, I
don't have a sampler so I'm limited to what I can sample,
hopefully the transmitter has a sample port
|
External Clock board for 332xx series
Hi fellows,
?
Some years ago I made a external clock input board for my HP/Agilent 33220A.
?
I offer two remaining PCB¡¯s few weeks ago on this site and they immediately gone.
My question is, whether there is any more interest in this or to archive this project?
On my side, to make it cost effective, more than 7 orders are required. The price would be 99USD including the shipping cost to US or Europe and will take about 4-5 weeks to build-up (holiday season coming¡).
?
Design are based on original Agilent 33220-80001 (33220-66503) board and compatible with 332xx series of function / arbitrary waveform generators, but slightly modified.
Input impedance is 50ohm AC coupled, DC isolated from instrument housing and sensitivity of -13 +/-2 dBm. Input level for optimal jitter/phase noise is between 0 and 20dBm. Board is only 10MHz input, no loop back output!
?
Picture of board installed in my 33220a is attached. New revision will have MCL or Coilcraft 1:1 input isolation transformer.
?
Regards,
Milan
|
Re: 8566 to measure 100W FM transmitter
What are you trying to measure, the fundamental, modulation, harmonics or what? #this will determine what bandwidth and accuracy you need
Is the measurement off-air or into a dummy load. If a load a power attenuator is best. if off-air some kind of coupler will be required. a resistive tap like
? or
would do and is reasonably broad band.
If looking at harmonics you will probably have to put a notch filter in the sampled ignal to reduce the carrier level and prevent spurious signals being generated in th analyser.
Robert G8RPI.
|
Re: 8566 to measure 100W FM transmitter
First of all you must be very strong. The 8566 A or B is somewhat over 100 pounds and very bulky, if you have to take it to the transmitter. I have one of each and try to never move them.
I would keep the input power below one half watt on the 8566 input, to be safe. Never put in zero dB attenuator on the analyzer!
You will either need a large 50 ohm attenuator of at least 40 dB, with an additional low power 10 dB attenuator to sample the transmitter and protect the spectrum analyzer. That is what I have on the bench.
The other way to measure the transmitter is to use a high power directional coupler in addition to a small attenuator on the forward coupled arm, for insurance.
I routinely measure two 21 kW FM transmitter, but they have a built-in 55 dB directional coupler. However, I always use an additional attenuator at the spectrum analyzer, just in case.
If you are just looking at a SSB transmitter, at a very narrow dispersion, you probably can use just a test lead. If you are just interested in relative readings, such as close-in distortion, carrier or unwanted sideband suppression in a Single Sideband transmitter, that will work well.
Stuart K6YAZ
Los Angeles, California
|
Re: 8566 to measure 100W FM transmitter
Why not just use a foot
long piece of coax, any kind will do, make a pig-tail out of
the far end and attach an alligator clip to the center
conductor and just lay it on the bench. It will make a dandy
RCV antenna.
If you are any where
near the transmitting antenna and the transmitter is
outputting 100W of RF at VHF you will have plenty of RF signal
to sample. If need be bring along either an adjustable step
attenuator and a couple 20dB or 40 dB fixed attenuators in
case the level is still too strong.
I have done this dozens
of times to monitor my own RF signal at various frequencies
when I do not want to go to the trouble of connecting up a
directional coupler and a step attenuator. You do not need to
go to the trouble to capacitive couple to the actual
transmission line of the transmitter.
Gedas, W8BYA
Gallery at
Light travels faster than sound....
This is why some people appear bright until you hear them speak.
On 5/23/2019 2:56 PM, DW wrote:
toggle quoted message
Show quoted text
I have a 8566 and I need to sample the signal of a 100W FM
transmitter. I haven't seen the transmitter but I was told the
following below.
Take a piece of wire and attach it to the end of the cable that
will go to the 8566. One end will go to the center conductor and
the other will go through a resistor to the shield of the cable.
This wire will loop once around the transmission line. So far I
made two various wires, one with a 400 ohm resistor and another
with a 200 ohm resistor. I will take a 2W attenuator with me for
good measure. Any thoughts on this, thanks. I ask this as I'm
being cautious about the input level to the 8566 as I know it
doesn't take much to destroy the front end of a spectrum analyzer.
Thanks
|
Re: 8566 to measure 100W FM transmitter
I built one of these:
You might want to use a larger resistor and make it a 50 dB tap. I use that for my amp output. Of
course you have to calibrate it using a SA and TG or a VNA because the
attenuation is not exact, but I have had good results through HF feeding
the output to a SA or 50 ohm scope input.
Regards,
Mark W7MLG
I should add the 100W will be driven to an antenna on a tower, I don't have a sampler so I'm limited to what I can sample, hopefully the transmitter has a sample port
|
Re: 8566 to measure 100W FM transmitter
I should add the 100W will be driven to an antenna on a tower, I don't have a sampler so I'm limited to what I can sample, hopefully the transmitter has a sample port
|
Re: 8566 to measure 100W FM transmitter
I believe it is 99.1mhz I have a 90dB manual step attenuator so that should help some along with a resistor to help pad down the levels, how hot do sample ports usually run on small FM transmitters I wonder. Thanks for the replies
|
Re: 8566 to measure 100W FM transmitter
Do you need to measure it on the air? What frequency is the Tx?
Is there a sample port on the transmitter?
Regards
On 5/23/2019 2:56 PM, DW wrote:
I have a 8566 and I need to sample the signal of a 100W FM
transmitter. I haven't seen the transmitter but I was told the
following below.
Take a piece of wire and attach it to the end of the cable that
will go to the 8566. One end will go to the center conductor and
the other will go through a resistor to the shield of the cable.
This wire will loop once around the transmission line. So far I
made two various wires, one with a 400 ohm resistor and another
with a 200 ohm resistor. I will take a 2W attenuator with me for
good measure. Any thoughts on this, thanks. I ask this as I'm
being cautious about the input level to the 8566 as I know it
doesn't take much to destroy the front end of a spectrum analyzer.
Thanks
|
Re: 8566 to measure 100W FM transmitter
the best Thing woud be to use a power attenuator e.g. 40 dB /500 W plus another 10 dB attenautor between the? Output of the poer att and Input of the SA. Remember 100 W = +50 dBm, so an overlal attenuation of 50 dB (40 dB + 10 dB) will give You 0 dBm = 1 mW at the Input of Your SA, save enough for Yoru Instrument.
Henning Weddig
-----Urspr¨¹ngliche Mitteilung-----
Von: DW <wilson2115@...>
An: HP-Agilent-Keysight-equipment < [email protected]>
Verschickt: Do, 23. Mai 2019 20:56
Betreff: [HP-Agilent-Keysight-equipment] 8566 to measure 100W FM transmitter
I have a 8566 and I need to sample the signal of a 100W FM transmitter. I haven't seen the transmitter but I was told the following below.
Take a piece of wire and attach it to the end of the cable that will go to the 8566. One end will go to the center conductor and the other will go through a resistor to the shield of the cable. This wire will loop once around the transmission line. So far I made two various wires, one with a 400 ohm resistor and another with a 200 ohm resistor. I will take a 2W attenuator with me for good measure. Any thoughts on this, thanks. I ask this as I'm being cautious about the input level to the 8566 as I know it doesn't take much to destroy the front end of a spectrum analyzer. Thanks
|
Re: WTB AMI/Mostek MK6220N RAM Chip
Adrian I guess only time will tell. But when I was getting the chip to temporarily work with heat, I barely put the tip of the soldering iron for more than 2-3 seconds on a pin solder junction, not really enough to be able to heat up any of the stuff that was underneath. Or maybe there was a bit between two adjacent pins, but I recall very briefly applying heat to different pins and always getting the same result. Now the very last thing left is to fix is the display, which has a faulty a-segment on the last right most digit. I could live with it, but it would be nice to see it whole again.
Alex
|
8566 to measure 100W FM transmitter
I have a 8566 and I need to sample the signal of a 100W FM transmitter. I haven't seen the transmitter but I was told the following below.
Take a piece of wire and attach it to the end of the cable that will go to the 8566. One end will go to the center conductor and the other will go through a resistor to the shield of the cable. This wire will loop once around the transmission line. So far I made two various wires, one with a 400 ohm resistor and another with a 200 ohm resistor. I will take a 2W attenuator with me for good measure. Any thoughts on this, thanks. I ask this as I'm being cautious about the input level to the 8566 as I know it doesn't take much to destroy the front end of a spectrum analyzer. Thanks
|
Obviously I took on more than I can handle. I was hoping a few interested fellows would come over and we could work out some deals. That does not seem to be happening. I am overwhelmed by emails, only a few with offers and the others asking how much, even though I asked for offers. Also, most are interested in the smaller items like horns and so on. This essentially means that I need to go through all of the small stuff to find specific requests. If I am going to do that, I might as well start selling items as I go through them. More than likely I will strip all of the heavy stuff and sell the essential pieces. So, here are three items for sale to start that are easily mailable. ? HP Harmonic mixers ¨C ? 11970K $200 11970A $250 11970U $550 ? If you want all three, I will accept $800 plus shipping. I will also consider offers on the others. As soon as these sell, I will list other items or wait a week and list other items. Thanks & 73 ¨C Mike ? ? ? Mike B. Feher, N4FS 89 Arnold Blvd. Howell NJ 07731 848-245-9115 ?
toggle quoted message
Show quoted text
From: Mike Feher <n4fs@...> Sent: Tuesday, May 21, 2019 2:16 PM To: ' [email protected]' < [email protected]> Subject: TE FS - ? Hi all ¨C ? Since I hardly ever use my lab any more, I decided to sell most of its contents. Unfortunately I am unable to ship heavier items due to physical limitations here. Consequently, if you would like to get something it is best if you pay me a visit. The more you take, the better the price, HI. Small items I can ship or even place on eBay. PP is fine with me. Any interest then please call or send me a direct email at n4fs@... . You can also try calling on the number below, however I do not have an answer machine set up on it. At the following link you can see 64 pictures, some not that great, taken inside the lab. In addition I also have a 2000 sq. foot basement, a 600 sq. foot garage and a 1000 sq. foot shed all full of stuff including loads of TE. Looking for offers until I place the smaller stuff on eBay. Thanks & 73 ¨C Mike ?
? Mike B. Feher, N4FS 89 Arnold Blvd. Howell NJ 07731 848-245-9115 ?
|
Re: WTB AMI/Mostek MK6220N RAM Chip
Ok, so my theory is that the stuff you cleaned up under the chip was hygroscopic and when damp was conductive enough to cause a problem.
Re-soldering the IC pins dried it out a bit allowing things to work for a while. Now you've cleaned it up it's good for the next 45 years....?
Adrian
toggle quoted message
Show quoted text
On 5/23/2019 1:56 PM, Alex wrote: Here is an update on the removal of the RAM chip from the main board.
The IC was unsoldered with soldapult pump and plenty of liquid flux. To clear some of the vias and since most pins where resting on the inner walls of the vias, had to first center them in the hole and then retry clearing it out of solder. On the top side of the board had to use solder wick to clear some excess solder holding some of the pins to the board. It took a while but eventually managed to gently pry the chip loose.
There was indeed still some corrosion left on and between 3 of the 4 traces that run under the IC, but very light not really much. Cleaned up traces with 91% IPA, and a Q-tip. The IC body itself seemed clean on the bottom and in between the pins, but just in case gave it a gentle brushdown with IPA as well. Also cleaned up the residual flux left on the board. Happned to have two 10-pin low-profile socket strips (these cost almost 50 cents/pin!) so trimmed then down to 8 pins each and mounted the RAM chip to align all the pins properly and then soldered the socket. Only soldered the pins that are actually in use which on this chip is only 8 out of 16. Just in case to make it easier if in the future I ever needed to remove the socket.
After powering up the calculator it worked 100%. It certainly must be due to residual heat from the desoldering/resoldering operation.
30 minutes later it still works. Completely removed power and let it sit for about 3 hours, test again: still working. Put the board back into the calculator case and reassembled it, put in the battery pack: still going.
It is now almost 20 hours after the repair and the calculator is still working 100%. Last night gave it a through workout to test all functions, and ran a short program. Everything worked.
Only question is, will this repair really last? Or was it the residual corrosion that was probably partially shorting some of the traces. Or did the added heat of the removal and reinstall of the RAM chip have a more lasting healing effect? Anyway, as it is now seems fine for now but I will keep an eye on it.
Anybody care to comment on why apparently this RAM chip came back from the dead? Or was it really dead to start with?
Alex
|
Re: WTB AMI/Mostek MK6220N RAM Chip
Once I took a flash photo (cameras used to have a flash that used special bulbs with magnesium filaments, and the camera spat out a picture on paper that appeared in about a minute) of a friend as he worked on a, probably 8080 or Z80, computer circuit that he had just got working. Immediately his circuit released a cloud of smoke before he could turn it off. After repairing the power supply the circuit still did not work and it was time to go home. The next morning the circuit was back to working. The cause was an open EPROM window and apparently the flash turned on elements that shorted the power supply. After a rest the chip recovered. I have seen other chips recover from abuse but replace them if possible. Yours may last another 20 years.
Here is an update on the removal of the RAM chip from the? main board.
The IC was unsoldered with soldapult pump and plenty of liquid flux. To clear some of the vias and since most pins where resting on the inner walls of the vias, had to first center them in the hole and then retry clearing it out of solder. On the top side of the board had to use solder wick to clear some excess solder holding some of the pins to the board. It took a while but eventually managed to gently pry the chip loose.
There was indeed still some corrosion left on and between 3 of the 4 traces that run under the IC, but very light not really much. Cleaned up traces with 91% IPA, and a Q-tip. The IC body itself seemed clean on the bottom and in between the pins, but just in case gave it a gentle brushdown with IPA as well. Also cleaned up the residual flux left on the board. Happned to have two 10-pin low-profile socket strips (these cost almost 50 cents/pin!) so trimmed then down to 8 pins each and mounted the RAM chip to align all the pins properly and then soldered the socket. Only soldered the pins that are actually in use which on this chip is only 8 out of 16. Just in case to make it easier if in the future I ever needed to remove the socket.
After powering up the calculator it worked 100%. It certainly must be due to residual heat from the desoldering/resoldering operation.
30 minutes later it still works. Completely removed power and let it sit for about 3 hours, test again: still working. Put the board back into the calculator case and reassembled it, put in the battery pack: still going.
It is now almost 20 hours after the repair and the calculator is still working 100%. Last night gave it a through workout to test all functions, and ran a short program. Everything worked.
Only question is, will this repair really last? Or was it the residual corrosion that was probably partially shorting some of the traces. Or did the added heat of the removal and reinstall of the RAM chip have a more lasting healing effect? Anyway, as it is now seems fine for now but I will keep an eye on it.
Anybody care to comment on why apparently this RAM chip came back from the dead? Or was it really dead to start with?
Alex
|
