Manual is online: google keyword "HP 339a" gets this link to the operating and service manual: http://cp.literature.agilent.com/litweb/pdf/00339-90001.pdf
The manual above shows you how to re-calibrate, etc.
Less than $160 !! A killer deal if it works at all. ( Factory referb and calibrated = $1600 )
The manual above shows you how to re-calibrate, etc.
Less than $160 !! A killer deal if it works at all. ( Factory referb and calibrated = $1600 )
Last edited:
Hi TubeMack,
You got an amazing deal! I love my 339A and it cost me $400 ish - not working properly. I repaired it and promptly replaced my older Leader unit. No looking back at all. The 339A I'd love to get would be an option 001. And yes, the prices on Eeekbay fell considerably. I should have waited.
Currently, I'm investigating how to improve the performance of the 339A. I figure that improving the power supply and bypassing, and also the oscillator section might get me where I want to go. Anything I do will be done very carefully. This unit was designed by people far more intelligent than I.
-Chris
You got an amazing deal! I love my 339A and it cost me $400 ish - not working properly. I repaired it and promptly replaced my older Leader unit. No looking back at all. The 339A I'd love to get would be an option 001. And yes, the prices on Eeekbay fell considerably. I should have waited.
Currently, I'm investigating how to improve the performance of the 339A. I figure that improving the power supply and bypassing, and also the oscillator section might get me where I want to go. Anything I do will be done very carefully. This unit was designed by people far more intelligent than I.
-Chris
Yes, I'm happy! So far I have done no more then test the OSC. Looks like a have the dreaded dirty switch issue. Looking at the trace on the scope shows an eratic trace. It jumps around, cuts out. Breaks down into sputtering streamers. Messing with the dials will sometimes fix it, and sometimes cut it out altogether. Sounds like a cleaning is in order no? I have also noticed that the small lower Freq verner control just barely moves the traces a Div or 2. Is this right. It'd hard to compare to the "middle" dial up above, as the upper one is so eratic right now and throws the trace around so much.
I peaked inside, and those wafer switches are going to be a bear to clean. It will be hard to see whats going on due to the compactness!
While in, I took the time to check that Knob sloppiness of the Osc level control. Hmmm, it's appears unsupported up front, as the shaft just goes through a bigger hole in the panel? Wheres the support? How sloppy is this control on your units? Looks like thats how it's was designed?????
Thats it for now!
Last edited:
Hi TubeMack,
Yes, mine is sloppy as well. Not too bad, and I can see that this isn't a big deal really. In fact, I had completely forgotten about this, and I use it most days. You will also forget about this in time.
The "dreaded dirty switch issue" isn't an issue. It's age and expected. It's not too difficult to correct. Although it does take time and patience to do, just relax and calmly go about cleaning when you are alert. Take breaks along with a coffee - definitely not a beer or other alcoholic beverage! While you are working away, examine the rest of the unit calmly and make sure all shields are secured in place. Don't over tighten any screws. Stay away from the variable capacitor in the shield, and do not attempt to clean it. Moving any wire or component will affect the calibration, and you do not want to go there. You may know this, but others will read the thread also.
The first thing you must not do is attempt to remove any knobs! Some may come off okay, but they are easy to break and the lettering comes off very easily. Take care when cleaning the front panel when working around any switches. I made this error once, and inherited other people's "errors" as well. What happens is that the set screws become corroded to the shaft. Attempting to release the set screw results in the socket portion cracking apart. Nice, now it's a threaded stud.
To clean, you shouldn't try to use any type of spray cleaner at all. There are a number of trimmer capacitors, and if you get fluid in there, you're toast! What you should be using is cleaning solution designed to apply with a dropper or needle type tool. If you cut a paper "Q-Tip" (not the plastic kinds!!!), you can use it to apply the cleaner directly to the slip ring. Do not attempt to clean the contacts, but you can place a small amount of cleaner by the contacts themselves. Allow the cleaner time to do it's job, then very lightly, use the same paper part to rub the slip ring. You will find that the oxides come off easily and the surface of the slip ring should regain it's normal brass coloration. Don't worry if there are stains on the metal surface when you are done. I tend to pull most of the cotton off the "Q-Tip" (cotton swab) and use that to do the same job. But you have to be extremely cautious and watch the cotton does not catch any of the fingers that contact the slip ring. If this happens, you may twist the contacts out of shape. Then you have a serious problem on your hands. I also find it easier to hold the cleaning surface in place and rotate the switch while rotating the cotton bit. This keeps the cotton wound tightly and reduces any chance of catching a finger. You will go through a number of these. Just like cleaning a gun barrel, you keep cleaning until the cotton comes out clean instead of the black to dark brown it is coloured when you first start. Once the surfaces are clean, you can apply a lubricant. I have heard of people using "Vaseline", but I've never tried that.
What cleaner to use? I have used "WD-40" on occasion with badly oxidized contacts and slip rings. I normally use something called "De-Ox-Id" by GC (General Cement, 10-1902). Any similar cleaner should do the trick. Stay completely away from tuner cleaner.
Recognize that the main points are to apply only enough cleaner to work on the contact surfaces. There should never be any that drips onto the PCB or surrounding components - ever! Finally, you can use some fine, single weight machine oil with no additives to lubricate the bearings on the switches or controls. Use a solder "poker" or other fine pointed metallic object to apply the oil in small drops directly to the shaft / bearing contact points. Once again, do not allow any oil to drip onto the PCB or components around where you are working. All the switches should now operate smoothly and the contacts should be reliable. My 339A is now two years older after I cleaned it up. It works like a new unit and I've had zero contact problems. So much for the "dreaded dirty switch issue". It's now a thing of the past.
The last thing to check before putting your 339A into service is the condition of the power supply capacitors. Use a 'scope to check how much ripple there is, and the shape of that waveform. If it looks like the standard ripple, your capacitors are fine. If you see a "pip" on the leading edge of the ripple waveform, the capacitor needs to be replaced. The negative supply will have an inverted waveform - just for the sake of completeness.
There you go, your 339A should be serviceable. That is unless it has a fault or excessive distortion. If you connect the oscillator output directly to the analyzer input, you should be able to measure the distortion. On every unit I have checked so far (three), the distortion reading has been 0.0018% or very slightly less. Allow time for the auto nulling feature to do it's job as you switch to higher sensitivity ranges. The other performance check is to confirm the accuracy of the basic AC voltmeter calibration. Again, on each 339A I have checked, the AC voltmeter has been dead on the money. It's a true RMS responding meter, using the AD536AJ RMS detector IC. Against my HP 34401A, these AC voltmeters are dead on the money. So as long as the AC voltmeter is accurate, and the filter is reaching it's proper null, your 339A should be working properly. If you are using any digital AC voltmeter less accurate than an HP 3478A, 3457A, 3458A (I wish!!) or anything in the 34401A family, do not attempt any adjustments. These meters hold their calibration extremely well, so a reading error will probably be the digital meter, not the 339A. If it's out of calibration, send it into an Agilent calibration lab, or any other calibration facility using modern equipment where the standards are calibrated at least once a year, two years for the better calibrators (5500A or better, possibly a 5700A). The trick isn't getting the base low frequency to read right. It's the frequency response. As long as no one has been playing inside your 339A, it should still be in tolerance.
-Chris
Yes, mine is sloppy as well. Not too bad, and I can see that this isn't a big deal really. In fact, I had completely forgotten about this, and I use it most days. You will also forget about this in time.
The "dreaded dirty switch issue" isn't an issue. It's age and expected. It's not too difficult to correct. Although it does take time and patience to do, just relax and calmly go about cleaning when you are alert. Take breaks along with a coffee - definitely not a beer or other alcoholic beverage! While you are working away, examine the rest of the unit calmly and make sure all shields are secured in place. Don't over tighten any screws. Stay away from the variable capacitor in the shield, and do not attempt to clean it. Moving any wire or component will affect the calibration, and you do not want to go there. You may know this, but others will read the thread also.
The first thing you must not do is attempt to remove any knobs! Some may come off okay, but they are easy to break and the lettering comes off very easily. Take care when cleaning the front panel when working around any switches. I made this error once, and inherited other people's "errors" as well. What happens is that the set screws become corroded to the shaft. Attempting to release the set screw results in the socket portion cracking apart. Nice, now it's a threaded stud.
To clean, you shouldn't try to use any type of spray cleaner at all. There are a number of trimmer capacitors, and if you get fluid in there, you're toast! What you should be using is cleaning solution designed to apply with a dropper or needle type tool. If you cut a paper "Q-Tip" (not the plastic kinds!!!), you can use it to apply the cleaner directly to the slip ring. Do not attempt to clean the contacts, but you can place a small amount of cleaner by the contacts themselves. Allow the cleaner time to do it's job, then very lightly, use the same paper part to rub the slip ring. You will find that the oxides come off easily and the surface of the slip ring should regain it's normal brass coloration. Don't worry if there are stains on the metal surface when you are done. I tend to pull most of the cotton off the "Q-Tip" (cotton swab) and use that to do the same job. But you have to be extremely cautious and watch the cotton does not catch any of the fingers that contact the slip ring. If this happens, you may twist the contacts out of shape. Then you have a serious problem on your hands. I also find it easier to hold the cleaning surface in place and rotate the switch while rotating the cotton bit. This keeps the cotton wound tightly and reduces any chance of catching a finger. You will go through a number of these. Just like cleaning a gun barrel, you keep cleaning until the cotton comes out clean instead of the black to dark brown it is coloured when you first start. Once the surfaces are clean, you can apply a lubricant. I have heard of people using "Vaseline", but I've never tried that.
What cleaner to use? I have used "WD-40" on occasion with badly oxidized contacts and slip rings. I normally use something called "De-Ox-Id" by GC (General Cement, 10-1902). Any similar cleaner should do the trick. Stay completely away from tuner cleaner.
Recognize that the main points are to apply only enough cleaner to work on the contact surfaces. There should never be any that drips onto the PCB or surrounding components - ever! Finally, you can use some fine, single weight machine oil with no additives to lubricate the bearings on the switches or controls. Use a solder "poker" or other fine pointed metallic object to apply the oil in small drops directly to the shaft / bearing contact points. Once again, do not allow any oil to drip onto the PCB or components around where you are working. All the switches should now operate smoothly and the contacts should be reliable. My 339A is now two years older after I cleaned it up. It works like a new unit and I've had zero contact problems. So much for the "dreaded dirty switch issue". It's now a thing of the past.
The last thing to check before putting your 339A into service is the condition of the power supply capacitors. Use a 'scope to check how much ripple there is, and the shape of that waveform. If it looks like the standard ripple, your capacitors are fine. If you see a "pip" on the leading edge of the ripple waveform, the capacitor needs to be replaced. The negative supply will have an inverted waveform - just for the sake of completeness.
There you go, your 339A should be serviceable. That is unless it has a fault or excessive distortion. If you connect the oscillator output directly to the analyzer input, you should be able to measure the distortion. On every unit I have checked so far (three), the distortion reading has been 0.0018% or very slightly less. Allow time for the auto nulling feature to do it's job as you switch to higher sensitivity ranges. The other performance check is to confirm the accuracy of the basic AC voltmeter calibration. Again, on each 339A I have checked, the AC voltmeter has been dead on the money. It's a true RMS responding meter, using the AD536AJ RMS detector IC. Against my HP 34401A, these AC voltmeters are dead on the money. So as long as the AC voltmeter is accurate, and the filter is reaching it's proper null, your 339A should be working properly. If you are using any digital AC voltmeter less accurate than an HP 3478A, 3457A, 3458A (I wish!!) or anything in the 34401A family, do not attempt any adjustments. These meters hold their calibration extremely well, so a reading error will probably be the digital meter, not the 339A. If it's out of calibration, send it into an Agilent calibration lab, or any other calibration facility using modern equipment where the standards are calibrated at least once a year, two years for the better calibrators (5500A or better, possibly a 5700A). The trick isn't getting the base low frequency to read right. It's the frequency response. As long as no one has been playing inside your 339A, it should still be in tolerance.
-Chris
I snagged a HP 3478A from Boeing Surplus a few years back, for $25. Seems to agree well with my several standard resistors and DC voltage standard, and I hope it's still accurate on AC as well. Be nice to have an external AC voltage standard, but it's good to know this meter holds up pretty well over time.
Chris, Thanks for the info. I had been reading all the other good info from you on this unit, throughout the thread, and was a big reason why I decided on the 339a. Any Idea who would carry the NON spray Deoxit locally? I'm ready to dig in again next weekend! Any other options besides q-tips? I've read warnings against using them on wafer swithes do to stuck fibers.
Hi frags,
Cool, but you don't have a front end, signal amplifier (not low levels), filters or protection. The 339A responds to very high frequencies (the harmonics) and your solution does not. The best place for your solution is hanging of the analyzer outputs of something like the 339A.
There is more going on with a distortion analyzer than first appearances. I have tried, and do not use my sound card for this. Plus, the computer ground is a noisy thing.
Hi TubeMack,
Well, try the electronics jobbers or on-line stores. This is a General Cement product (GC Chemicals), a liquid in a bottle. Some on-line stores probably have it as well. Check with Newark or Digikey. In Canada, you should find it at Sayal possibly, Electrosonic or Active Components.
Caig Laboratories may have a similar product as well. Just think how "green" this type of product is. Less used and no propellant. I only use a canned version if I can't reach the area. It's hard to deliver a short shot, you always get far too much out.
Hi Damon,
Great price!!! Let's just say that my 3478A meters cost me about 4 X more than yours. You might want to replace the Lithium battery before you lose your calibration constants. It takes a 2/3 AA, non-rechargeable with single lead on each end. I used the tab versions. You need one extra and a 10 K resistor to keep the memory alive as you transfer the batteries. That's my trick and it's saved me lot's on $$$.
-Chris
Cool, but you don't have a front end, signal amplifier (not low levels), filters or protection. The 339A responds to very high frequencies (the harmonics) and your solution does not. The best place for your solution is hanging of the analyzer outputs of something like the 339A.
There is more going on with a distortion analyzer than first appearances. I have tried, and do not use my sound card for this. Plus, the computer ground is a noisy thing.
Hi TubeMack,
Well, try the electronics jobbers or on-line stores. This is a General Cement product (GC Chemicals), a liquid in a bottle. Some on-line stores probably have it as well. Check with Newark or Digikey. In Canada, you should find it at Sayal possibly, Electrosonic or Active Components.
Caig Laboratories may have a similar product as well. Just think how "green" this type of product is. Less used and no propellant. I only use a canned version if I can't reach the area. It's hard to deliver a short shot, you always get far too much out.
Hi Damon,
Great price!!! Let's just say that my 3478A meters cost me about 4 X more than yours. You might want to replace the Lithium battery before you lose your calibration constants. It takes a 2/3 AA, non-rechargeable with single lead on each end. I used the tab versions. You need one extra and a 10 K resistor to keep the memory alive as you transfer the batteries. That's my trick and it's saved me lot's on $$$.
-Chris
Hi frags,
Hi Damon,
Great price!!! Let's just say that my 3478A meters cost me about 4 X more than yours. You might want to replace the Lithium battery before you lose your calibration constants. It takes a 2/3 AA, non-rechargeable with single lead on each end. I used the tab versions. You need one extra and a 10 K resistor to keep the memory alive as you transfer the batteries. That's my trick and it's saved me lot's on $$$.
-Chris
Thanks for the heads up on the battery!!Yes, it's not for professional purposes but for DIY it's a good solution. This sound card (Asus Xonar Essence) has ultra low-noise input -120db, 24bits resolution at 192kHz sampling rate (10Hz-90kHz response). I measured -118db noise for output+input, so computer noise is not an issue. The only thing it does not have is an input attenuator/protection. I'm using 10k potentiometer.
Last edited:
There are dynamic range constraints with a soundcard as the analyzer but with a little hep it works OK. This probe will help Sound Card Oscilloscope,Spectrum Analyzer Probe (audio) - eBay (item 120525264401 end time Feb-07-10 16:22:59 PST)
The unprotected input is the weakest point (making a toasted card all too possible). You do have issues with grounding that don't go away easily. Also finding software that isn't really clumsy is important. I use a soundcard with a package called Praxis from Liberty Instruments for all kinds of measurements. With a Juli@ card I can measure to 90 KHz, but I use a stand alone analyzer for normal work. Its easier, faster, has as much resolution and is much harder to break (and easier to fix).
The unprotected input is the weakest point (making a toasted card all too possible). You do have issues with grounding that don't go away easily. Also finding software that isn't really clumsy is important. I use a soundcard with a package called Praxis from Liberty Instruments for all kinds of measurements. With a Juli@ card I can measure to 90 KHz, but I use a stand alone analyzer for normal work. Its easier, faster, has as much resolution and is much harder to break (and easier to fix).
339A owners. I bought a 339A in good condition off ebay not long ago and for about $150 like Tube Mac but I haven't used it much. I found some bad solder joints inside that i fixed and the unit seems to work, but tell me if your 339A's do this...when I press the power switch the needle of my meter SLAMS to the right and then settles back down. is this normal? is this really the way this operated when new?
well i was thinking the same thing. that just cant be good for that needle and would the Brilliant minds at HP really design the thing like that?? My switches seemed to be ok from what i remember. I put it on the shelf last year and haven't used it thinking it was still not working correctly and i have just never dug it out again. be careful cleaning those wafer switches. I have heard the wafer sections can soak up the cleaning solution and cause leakage between switch sections.
Hi Damon,
Yes. Get it before it gets you! Save the calibration, because if you lose it, the meter will not be close to reading right. Remember, closed case calibration which means if the battery dies, those "electronic trimmers" reset.
Hi Demian,
Exactly!
I guess that's two of us who can do this either way and still use gear built for the task. The thing that I never could understand is that it takes a professional to use unprotected (easy to blow) equipment without damaging it, and yet it's always people who are not that familiar with these things that end up buying these items.
All,
When Demian said that your ground issues are still going to be a problem, did you understand why this is? Once you have two devices connected together, ground current may then flow. Especially with a computer involved!
I understand perfectly how strong the desire is to be able to make detailed measurements. Been there often enough myself. But purchasing things like expensive sound cards and putting them directly in harms way is not the way to go. You must build a front end for these in order to :
Now is a good place to mention that an analog THD analyzer makes a good front end for a sound card measuring system. A pair of HP 331A - 334A, one for each channel. That and the analog units will give you accurate readings to higher frequencies than the sound card. That's if you want to use a sound card.
Hi Zero Cool,
While we are on the subject of cleaning contacts, never use an eraser on things like edge connectors. The plating is easily damaged with an eraser.
-Chris
Yes. Get it before it gets you! Save the calibration, because if you lose it, the meter will not be close to reading right. Remember, closed case calibration which means if the battery dies, those "electronic trimmers" reset.
Hi Demian,
Exactly!
I guess that's two of us who can do this either way and still use gear built for the task. The thing that I never could understand is that it takes a professional to use unprotected (easy to blow) equipment without damaging it, and yet it's always people who are not that familiar with these things that end up buying these items.
All,
When Demian said that your ground issues are still going to be a problem, did you understand why this is? Once you have two devices connected together, ground current may then flow. Especially with a computer involved!
I understand perfectly how strong the desire is to be able to make detailed measurements. Been there often enough myself. But purchasing things like expensive sound cards and putting them directly in harms way is not the way to go. You must build a front end for these in order to :
- scale down large inputs
- provide protection for the input circuits
- amplify weak signals to be usable with a sound card
- maintain some kind of rough calibration
- keep from loading the circuit under test
Now is a good place to mention that an analog THD analyzer makes a good front end for a sound card measuring system. A pair of HP 331A - 334A, one for each channel. That and the analog units will give you accurate readings to higher frequencies than the sound card. That's if you want to use a sound card.
Hi Zero Cool,
Very odd, especially for an HP product. I have never seen one with a bad solder joint in any I've been into. Previous work possibly?
Yes, that's normal. It isn't as violent as you think and there is protection for the movement. Have you read the manual yet? As mentioned, the manual is available on www.agilent.com . Search for 339A.
Yes, but you have to look deeper to see what's really going on.
Sell it if you aren't going to use it.
Please read my instructions posted earlier. I was very clear about using a minimal amount of cleaning solution. Also, that wafer material is phenolic I think, and that's a compressed paper and wax structure.
While we are on the subject of cleaning contacts, never use an eraser on things like edge connectors. The plating is easily damaged with an eraser.
-Chris
I've not yet seen a bad solder joint in a piece of HP equipment.
According to friends in the T&E biz quite a long time ago -- it took HP a while how to figure out how to get an analog scope to trigger properly.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.