ad: Mircules-1

HP-432A Power Meter - On it's way to the grave!

Discussion in 'Homebrew and Kit Projects' started by AB8RO, Jul 7, 2008.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: L-MFJ
ad: Left-3
ad: L-Geochron
ad: abrind-2
ad: MessiPaoloni-1
ad: HRDLLC-2
  1. AB8RO

    AB8RO Ham Member QRZ Page

    So I hate making the kind of mistakes that I later regret, but, here goes.

    I have an HP-432A power meter with no thermister mount. As far as I can tell, that means I have one useless power meter in a nice little box. Judging from the following (and past) auctions they are worthless without the mount.

    In fact, as near as I can tell, the mount accounts for pretty much all of the sale price of any of these as they sell for as much or more than the meters with a mount.

    I could wait for a cheap mount, but, I already have a boonton 41A which, although isn't quite as sexy, has a probe that goes from 100khz to 7ghz and seems to work fine. If it doesn't work fine, well, there's no way I'd know because it's about the fanciest piece of test gear that I own.

    So, the plan is to rip out the guts, salvage a few parts like that very sexy 2N1701 transistor, and turn it into a combination inline wattmeter and dummy load. The cabinet size is fine for that and the front panel meter is nice to look at.

    The question is, is there any reason not to gut it? Is there any interesting use for the dual bridges that I'm not thinking of? (Since I can think of no use,that would be any use.)

    Edit : It just occurred to me that the power scale is linear on this thing making it useless for typical wattmeter applications.

    Edit: ARGH!!! Could one of the admins remove that pesky apostrophe from the title for me....thanks!
    Last edited: Jul 7, 2008
  2. VK2TIL

    VK2TIL Ham Member QRZ Page

    I have a mint 432A with two co-axial sensors (sorry, not for sale) and an "X-band" waveguide head.

    It's a beautiful instrument; I don't use it a lot as I have other power meters but I get it out if I need a "calibration check" as it's extremely accurate.

    But yes; no head and they are useless. The low tribo-electric effect cable is even rarer than the head.

    There are many more meters than heads because of the delicacy of the heads.

    So, much as I hate to see this happen, part it out and use what you can.

    The meter is probably the usual HP 1 mA type so a little insensitive for a low-power meter. FSD of 50 or 100 uA is usually preferred for low/medium power meters. It should be OK for a 100 watt or so meter/load. These HP meters are very rugged & accurate.

    If you want to try homebrewing, why not build an AD8307 power meter into the case?

    The W7ZOI/W7PUA version uses an analog meter and is easy to build; I understand that the QST article may be downloaded from the ARRL by members.

    Bob Kopski described a digital/analog version that is available as a download;

    Kanga does a kit. Google "AD8307 power meter" if you're interested. There are lots of homebrew ones.

    The meter scale is linear on these instruments; you might be able to just re-number the HP scale.

    The Boonton 41 is a good instrument but, much as I admire Boonton, it doesn't have the "quality" of the HP. Boonton were still learning how to tame diode detectors at that stage; they improved with the 42B (I have one of those with two "proper" heads and one homebrew head (not difficult) but, for sheer quality & accuracy, the 432A is hard to beat.

    Thermistors are still used today as standards where accuracy & stability are paramount.
    Last edited: Jul 7, 2008
  3. AB8RO

    AB8RO Ham Member QRZ Page

    Thanks, that looks good.

    Yeah, but if I want a 432 I pretty much have to just buy a new 432 with mount. Unless I'm willing to gamble that's about $300 tested/guaranteed. Joe blow gets around $100 or so it seems.

    My boonton cost me $10 and, for me, that's about right for a hobby.

    Thanks for the input.
  4. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    The 432A meter alone is pretty worthless, true. All the magic is in the mount, and the most popular mount (I think) is the 478A thermistor mount. I see Tucker Electronics has them for $350.00, which isn't a bad deal.

    I've picked up thermistor and bolometer mounts at Apex Electronics in Sun Valley, CA (drive-in local to me) for much less than this, though. I think I paid $150 for my last one, it was a 486A (similar to 478A).

    The Boonton 41A is very old but if it works, can't knock it. I have a model 42A, which is all solid state and years newer, but even it is c. 1974 (!). The huge advantage to the Boonton design is that most of the magic is actually in the meter circuit itself, and the detector is just a well compensated (and biased and chopped, using a signal from the main meter circuit) diode. You could almost homebrew the same thing for $50, the only magic is in the particular diode used and the microwave components used as a reference (50.00 Ohm microwave rod resistor) and decoupling (microwave chip capacitors). The diode was specially made for Boonton by ITT and I'll bet there aren't any available. They used to be kept in a safe at Boonton's Parsippany, NJ labs (where I was employed in the mid-70s) and only the President/CEO, VP of Engineering, VP of Operations and Engineering Manager had the combination. Seriously.

    But the 41/42 have some advantages, especially if dealing with fairly pure (sine wave, or close) signals: They were accurate from +13 to -50 dBm, 63 dB dynamic range, which is more than the HP's were. And they were not much influenced by heat. The heat transfered from a human hand to the HP thermistor mount would change calibration quite a bit, so when we used those, we had to wear gloves.


  5. VK2TIL

    VK2TIL Ham Member QRZ Page

    Homebrew a Boonton head, eh?

    I did this when I only had a 42B meter unit;


    I checked it up to a GHz and it worked as well as a Boonton one; I doubt that it would have matched the Boonton one at 12 GHz, though. :)

    I later obtained two Boonton heads.
  6. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    ::We had versions of the 50 Ohm RF detector head for the model 42 that were good up through 18 GHz with a small error curve that was printed on a label applied to the side of the head itself. The "magic dust" was just a good layout, good components and that special diode. The N connector was somewhat special (plated stainless steel to hold very tight mechanical tolerances).

  7. AB8RO

    AB8RO Ham Member QRZ Page

    If I'm reading that correctly it's a couple of 100 ohm resistors with the common point feeding a cap which goes into what I expect to be a pair of diodes which are then bypassed and connected to the probe inputs via 10k resistors?
    Last edited: Jul 8, 2008
  8. VK2TIL

    VK2TIL Ham Member QRZ Page

    Jolly Good Spotting!!

    Yes, that's about it.

    The 100-ohm resistors go to ground so determine a 50-ohm input impedance. The diode is a 3-legged HP Schottky type.

    It's just a dual-diode detector; the circuit is in the 42B manual.

    Getting something to work up to about 1 GHz is not too difficult with SMD components; above that, as WIK says, the "black magic" area begins. The 18 GHz N-connectors he refers-to would have been "precision N" types with unslotted mating sections; these are used at +12 GHz frequencies.

    The Boonton diode heads are surprisingly unsophisticated; they don't look as if they can go to 12 or 18 GHz. But they do indeed and very well. The diodes came with their matched load resistor attached; they were obviously matched for total series R at the factory.

    This bloke managed to repair one;

    In the factory, final "tweaking" was done by changing the position of the screws & offset washers that clamp the input load resistor pigtails to ground; this alters the parasitic inductance and thus "tunes" the input at GHz frequencies.
  9. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    ::That's also just about right. Sounds like you worked there, but I don't remember you. I wasn't on the 42A/B/C/D project but that group was right down the lab from me; I was on the 102A/B/C/D etc. signal generator, with Ray Polen W2WCF and others. We all worked for Ray Lafferty who was chief engineer at the time.

  10. VK2TIL

    VK2TIL Ham Member QRZ Page

    "Sounds like you worked there ......"

    Not me; never been in the US.
Thread Status:
Not open for further replies.

Share This Page

ad: Amateur-1