Fixing a HP200AB Audio Oscillator, where to start?

[rwgast_lowlevellogicdesin]

So i picked up this piece of gear a while back sitting a an e-waste place. I just downloaded the user manual/schematics. I would like to get it functioning properly again! I have an AWG, but this HP is nice because it outputs much higher voltages than a digital AWG.



It is in really nice cosmetic condition, but has multiple problems. The frequency is pretty far off from the set frequency on the dial. Sometimes I have to turn it on and back off and on again to get it working. The connections need a good cleaning and the Variable cap needs some DeoxIT. What I am most concerned about is the distortion, using the 600 ohm impendence posts on the front the wave looks pretty bad



If I plug my leads in to the black/red post I get an even worse waveform. 



I have never worked with tubes before and I am pretty unfamiliar with the issues that arise when a tube is bad, I'm not sure if they weaken of just die. The first waveform looks to me like it could possibly be a tube clipping the Sin wave, but I'm not even sure if that is a thing a tube would do when close to failure.

I went to take it apart and start testing components mainly caps, but these caps are nuts and I'm not sure the best way to get them out. I think these two bigger caps are non polarized and there the kind where you can open the tube and there is smaller ones inside, but I'm not even sure about that yet because I don't know a good way to get them out of the unit.

Here is the first cap



It looks like the only way to remove this guy is to drill out those rivets?

The second bigger cap doesn't even have riveting, I'm not sure but it looks like maybe you are suppose to twist the lead blades, and that is all that is holding it to the chassis???



If anyone on here has worked on one of these guys, or some HP gear of the same vintage and knows the best way to remove these caps I would very grateful. I watched some you tube videos but most are for the 200CD not 200AB and they never show how they actually remove the caps. For now I figure I would just go through the schematic cleaning and testing parts (there isn't not to many) and replacing anything bad, but I don't want to go drilling unnecessary holes to get parts out!

[coppercone2]

Well I have alot of Hp test equipment with those caps and I can tell you think

1) most people say if they test fine their fine
2) I cut a whole bunch apart with a band saw, that tested fine. The only thing I see is browning on some caps but others remain toilet paper white. They smell sweet inside like some kind of fruit chemical.


I replace them all. You use flat grip pliers to straiten out those tabs that are twisted in and they pull right out once smashed flat enough, and can be bent back into place.


You can either try to cut that cap apart and reuse its base to mount a capacitor or make some hardware. The bases kinda smell but glue does seem to stick to it after a good cleaning, so you can drill holes in there after cutting the tabs flush and possibly reinforcing them with glue. Top side inside of the can that you might remove, its aluminum thats like crimped to steel or some thing like that, its not very good for solder. I would recommend drilling 2 small holes in there so you can feed capacitor leads through and wrap them around the soldering tabs on the bottom of the bakelite substrate if you are hell bent on reusing stuff. I don't care for the original can, the new caps have the nice vent in them so they can fail gracefully, and you get more room in the chassis if you ever wanna do some mod.




The nice option however, is to drill the rivets, make a diamond shape PCB that fits over the existing holes, bolt that in place with a capacitor in the middle, and rivet two solder posts to the bottom so you can do your point to point wiring (feed capacitor leads through PCB to wrap it on the binding post). You could even get big threaded studs in the PCB, so you can just bolt it to the chassis without a nut. In this case you just need to select the right hardware and get a diamond shape pcb material cut out (i.e. if there is alot of wires going to the leads, get some big ass solder posts. The one that looks like a chirstmas tree made of disks.

Now if you actually get a plated PCB, you can solder the capacitor down and use traces to lead the capacitor to you feed throughs rather then using silicone and wire pulling to get the capacitor connected to your solder posts. What is important to get similar to original 'quality' is that you use riveted hardware for soldering, so you have nice heat resistant easy to solder and reliable connectors. You could also just solder the wire to a glued down capacitor on a PCB that fits over the original chassis hole, but its weak and looks bootleg


another option is to put the capacitor in the diamond hole after its cleaned, and put alot of JB weld around it to fix it in place near the crimped end so the putty confroms around it but leaves the bottom and top of capacitor undisturbed. This is the easiest solution if your cap can fit in that hole, the one ones are alot smaller, but you have to live with soldering wire to the capacitor leads without some kind of  durable interconnect. If you do this, for best attachment, you need to skin the capacitor to get the JB weld to glue to the metal can not the film on it, incase it wants to rotate.


If you have high voltage on those caps, I suggest making it more robust, if you desire to replace them. The browning I found made me a little nervous....


a cheap chinese distributor can make a PCB for you that has all the holes drilled, so you only need to rivet on the hardware and solder everything in place.. then you can just screw it in. Only challenge is that making sure all the hardware is what you imagined in your design, lots of work to do in selecting the right parts, but it will be legit. You could even get tabs cut into your PCB so you can attach the can if you carefully open it and reform it after gutting and cleaning it for cosmetic reasons (retaining the bent tab mounting method for the can).

[Stray Electron]

  That type of capacitor mounting was very common in the 1950s and 1960s and the early 1970s.  You will have to drill out the rivets to remove the first cap and you will have to unsolder the wires from the tabs on the second cap and then remove as much of the solder from the tabs as you can and then straighten the tabs and then remove the cap from the chassis.  Note, that since the tabs were mechanically attached to the chassis, it was VERY common to use the tabs as ground lugs, even if the connection was for an unrelated circuit. 

    You should look and see if the chassis of the unit is connected directly to one side of the AC line. A lot of the early equipment used a two wire AC cord (often unpolarized) so it was easy to have the chassis connected directly to the hot side of the AC line and that can be a real safety hazard. My old Halliicrafters radio is built that way and since I don't want to modify it, I only connect it to the AC power line through an isolation transformer.  The early HP equipment generally used a two wire power cord but as long as you used the OEM power cord it was polarized on both ends so it could not be reversed and apply a hot line to the chassis but there was another available power cord that was had the wiring reversed and if you used one of those cords then you could have the hot side of the line applied to the chassis.  The original 2 wire HP power cords are VERY hard to find and typical bring nearly $100 each on Ebay so my advice is that if you have that type power connector, and you don't have the correct power cord, then remove it and replace it with a later 3 wire power connector that uses a standard IEC power cord.

  Due to the difficulty and great expense of replacing those caps, I would disconnect the wires from them and thoroughly test them before deciding to replace them. HP usually used a very good grade of caps over the years and they really don't fail very often. My guess is that they're only filter caps for the rectified 60 Hz power so the caps aren't particularly stressed.

   There are about a million posts on the antique radio forums about how to replace or to "re-stuff" old capacitor cans so I won't get into that here. 

[factory]

There are only two large capacitors shown in the manual, the 4 section can in the PSU, measure DC voltages & AC ripple with your DMM? The slightly smaller can is the non-polarised one.

This is a 1960s or later 200AB, it will have a three cord power lead or connector, do however check the ground/earth connection, I've come across these with a bad ground connection due to the use of a rivet (only) for the tag strip ground.

Something that has not been mentioned is the fixed paper capacitors listed in the manual, i.e. the bumble bee or black beauty capacitors, please take some better pictures so we can see if it's using these paper axial capacitors, or later plastic dielectric capacitors.

David

[Stray Electron]

   I think that HP switched over to the 3 pin AC power connector in about 1965.  If the OP wants to post the SN of his unit I can tell him when it was made.

   The Bumble Bee caps and Black Beauty caps are notorious for failing due to age. I would expect that they're all going to need to be replaced.  Those Allen Bradley carbon comp resisters also drift out of spec with age. 

[factory]

The SN won't tell you the date it was made, only the date of the last major design revision (for post 1960 units only).

David

[ardiesse]

The heavily clipped waveform at the output terminals means that the Wien bridge is out of balance - the amplitude limiting function is not working.

The original HP manual will have a description of the theory of operation of a Wien bridge oscillator.  Pay particular attention to the section on amplitude stabilization - the HP 200s use a light bulb for this.  I'd also suggest that you try to track down a copy of Fred Terman's textbook "Electronic and Radio Engineering".  As well as a detailed explanation of the Wien bridge oscillator, Terman's book has much material on the fundamentals of vacuum tube circuits.  You need to understand how a Wien bridge oscillator works before attempting repair work.

In equipment like this, the valves (sorry, tubes) are in general the most reliable components.  The oscillator produces an output, so you can assume they're all in working condition.  And hold off replacing any capacitor, including the electrolytics, until they are shown to be bad.

If you've never worked on valve equipment before, beware of the high-tension supplies.

Rob

[rwgast_lowlevellogicdesin]

I read through the manual

http://hparchive.com/Manuals/HP-200AB-7725-Manual.pdf

and it says for waveform distortion bulb to investigate bulb R11, so ardiesse is right, it is a problem in the wien bridge oscillator. The manual says it should be a 250v 10watt, the bulb I pulled out is a 120v 7w bulb made in China and looks fairly new.



the thing is I don't know how to get a replacement bulb. I mean I know I just need a 250v 10w incandescent bulb, but I don't know what kind of socket that is, it is much smaller than a regular household fixture. I'm in the US on 120 mains what is the likely hood of finding the right kind of bulb at home depot or something? The current bulbs filaments are fine and it measures about 250ohms, it does not light up when installed in the circuit, I'm guessing it is not suppose to though?

[ardiesse]

A 250V, 10W light bulb will have a nominal resistance of 250x250/10 ohms, which is 6.25 k.  But the "hot" resistance is much higher than the resistance at room temperature - this is how the amplitude stabilization works.  In operation in a Wien bridge oscillator, the filament will glow red or dim orange, and in a well-lit room, you might not be able to see it glow.

Your 120V, 7W light bulb will have a "hot" resistance of 120x120/7, about 2 k.  That's a factor of eight higher than its resistance at room temperature (250 ohms).

At a guess, the bridge will probably balance for a filament resistance between 500 and 1000 ohms.

You're in luck.  Digi-Key part number 10S6-230V (and in stock in the USA, too) is a 230V 10W candelabra screw light bulb.  I bet it'll be a drop-in replacement.  Maybe even your local lighting store has them.

[rwgast_lowlevellogicdesin]

Thank you! Candelabrea bulb is exactly the information i needed!