Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 2 [3]   Go Down
  Print  
Author Topic: Question on probe for VTVM  (Read 20999 times)
KB1WSY
Member

Posts: 804




Ignore
« Reply #30 on: April 22, 2012, 05:21:02 AM »

So I finally disassembled my old Eico 232 VTVM (dated June 1962) and rebuilt it almost entirely from scratch, using the original kit instruction manual that came with the unit (it still has the penciled check marks from 50 years ago!). I replaced all the capacitors, removed extensive corrosion from the battery holder and chassis, and replaced the power cord, the plastic shoulder washers for the battery holder, and most of the old hookup wire. The only thing I left alone was the range switch with all the precision resistors soldered to it since it was in fine condition. I nervously checked all of my wiring against the instructions and found "only" one mistake which I fixed before powering up.

Surprise! It worked! Only one problem: both of the DC voltage options (DC- and DC+) were intermittent -- they sometimes stayed at zero volts regardless of the applied potential. I then found and removed some mild corrosion on the spring switch inside the Uniprobe, and I also slightly increased the flexing of the springs, and I now have a consistent 1 megohm resistance in the probe's DC volts position, rather than the occasional open circuit. But the original problem didn't go away entirely. The "DC-" range now works properly all the time but the DC+ range is still mostly dead. I checked all the voltages inside with a digital VOM (B+, bias, etc.) and everything is perfect. Conclusion: it must be a problem with rotary function switch S1. This is confirmed by the fact that I can sometimes fix the problem by rotating the switch a few times.

Space is tight inside there, but visually, the switch looks OK albeit with substantial wear on the contacts/metal rotors. Unfortunately it is one of those highly customized switches with "made to order" wafers, so replacing it with an off the shelf component isn't really on the cards. Which leaves the choice of (1) finding another junked Eico 232 and cannibalizing it just for the switch, or (2) trying to fix the switch. Does anybody have advice on how to proceed with choice (2)? Should I try, for instance, DeOxit spray? They are phenolic wafers BTW, not ceramic. I tried a generic "electronic cleaner" spray but it actually made things worse, not better -- I think it put a semi-insulating coating on the contacts, rather than cleaning them. Meanwhile I am spending quite a lot of time doing some electronic sleuthing with the circuit diagram and my digital VOM, trying to pin down *which* set of contacts is the culprit (this is harder than it sounds, because of the confined spaces, the complexity of the switch, and the fact that it is still connected to the rest of the unit, which introduces unrelated resistances into my simple "continuity testing").

The unit is perfectly usable even for DC volts, but I am confined to the DC- range, which is OK as long as I reverse the probes when measuring positive potential! The other ranges (AC Volts, Ohms) work beautifully and are very accurate, tallying closely with my $4.99 digital VOM.

Now, time to start on rebuilding my Eico 315 RF signal generator and I'm looking forward to using my "new" VTVM during that project. The fun never ends. By the way, if anyone has an old 315 dial face they don't want please let me know. My unit is in good condition except that the glass face on the dial is shattered. (It is a custom piece of glass with a painted red indicator mark on it for the vernier sub-dial, not too easy to homebrew/duplicate.)

73s DE Martin, KB1WSY


Logged
G3RZP
Member

Posts: 4715




Ignore
« Reply #31 on: April 22, 2012, 06:23:24 AM »

Martin,

Most rotary switch functions can be made up  by wiring either 'short all except the one you want' or  'select the one you want' wafers. But it can get big, needing several wafers. One trick I use is to buy up old wafer switches from the boxes of junk you see at hamfests. Then you can drill out old rivets, collect contacts and rebuild switches using 2-56 screws to hold contacts in place. If you really want to get clever, get some of the small eyelets used by craft people and use them as rivets. Careful work with pliers will normally allow you get contacts off rotors, and these can often be wide enough that you can make custom rotors. I did this when I rebuilt an old Heathkit VTVM.

We used to have great fun designing rotary switches to give the functions needed with the minimum number of contacts and wafers. One thing I would suggest you avoid if at possible are 'insulated clips' where the contacts front and back of a wafer in one position need insulating from each other.

There are basically 4 types of fixed contact, or 'clip' as they were known here. These were long clips - which contact the rotor wherever it is pointing, and short clips which ar the one the rotor contacts in each position. Both sorts are then either 'normal' or 'reversed'. A 'normal' clip has the contact bent such that the bit with the hole for the wire in it points to the same side of the wafer as the contact is on: the 'reversed' clip is the opposite.

Be careful when redoing rotors: sometimes you need a 'break before make' and sometimes a 'make before break'. The former usually have a narrow tongue on the rotor and the latter a wide tongue. Where possible, a wide tongue has advantages of allowing a bit more tolerance on angular positioning. But it can lead to a lower breakdown voltage.

Any input switch on a VTVM that's phenolic shouldn't be pushed past 500 volts, al;though the Heathkit one I rebuilt had a 1500 volt range - on a wafer switch and a banana jack input!
I can hear AC5UP's snort of derision, and he's certainly justified in that case.

I'd try DeOxit first, as being much easier. But collect old wafer switches at hamfests - and Dayton is coming up!
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #32 on: April 22, 2012, 07:21:33 AM »

I fixed it! My post was written after many, many hours of failed troubleshooting. Then I realized belatedly that the circuit diagram that came with the unit was wrong! Eico clearly changed its switch layout at one time or another. I found a more "recent" one at BAMA (dating to 1964, and my VTVM, dated June 1963, apparently shipped with an outdated manual or at least the one I was given with the unit was outdated!). Then I made up short "jumper cables" with alligators at both ends, and used them on the switch terminals to "duplicate" the switch circuits that were supposed to be be created by the switch, when in the DC+ position. I finally found that it was the most inaccessible wafer and its most inaccessible contacts that were the culprit: the very first wafer, at the front nearest the panel, and underneath the switch as viewed from the servicing position. I had to pull out the already wired switch (aargh) and then I gently pushed down on the top of the defective stator contact with a flat blade screwdriver to close the gap slightly. The VTVM is now working absolutely perfectly! I will also get some DeOxit and clean the switch. I wouldn't be surprised if this problem comes back later, but at least I know how to fix it!

Thank you G3RZP for the advice!

73 DE Martin, KB1WSY
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #33 on: April 24, 2012, 09:59:30 AM »

So, another dumb question from yours truly. Earlier posters advised me to ground the VTVM's case, replacing the original 2-wire AC cord with a new 3-wire one. I haven't done that yet. Part of the procedure for calibrating the voltmeter, as per the original manual, is to connect the VTVM across the house AC supply and then adjust the AC calibration pot so that the meter reads 117 VRMS. Which I did, with some trepidation: I was very careful not to touch the VTCM's metal case while doing this because one of the two probes is connected directly to the chassis and the chassis is therefore "hot" with 117 VAC while this measurement is being made. While doing this, it occurred to me that if the case had been grounded, the result could have been spectacular i.e. connecting one side of the house AC supply directly to house ground.

So my thinking is: either leave things as they are, keeping the 2-wire AC cord, but be very, very careful when measuring on the AC range to avoid electrocution and/or tripping the house circuit breakers. Or, ground the case but make sure that all AC voltage readings are taken at points that have already gone through an isolation transformer, e.g. the HT secondary side of a power transformer or the filament windings. Never connect directly to house supply or to the primary of a power transformer. Even when measuring secondaries, beware of situations where neither of the two AC secondary wires is connected to the chassis.

Correct?

TNX ET 73 DE Martin, KB1WSY
Logged
K8AXW
Member

Posts: 3900




Ignore
« Reply #34 on: April 24, 2012, 11:03:46 AM »

Martin:

I can't comment on your power connections, etc. because I'd need to see a schematic. 

However, the one thing I would like to comment on is the calibration procedure.  Setting the meter to 117VAC is pretty strange.  My line voltage varies from 115VAC to 125VAC, depending on the time of year, time of day and even the temperature of the day!

You need something else to measure your line voltage and then set you meter to that voltage and then it will be "calibrated" for AC.
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #35 on: April 24, 2012, 11:49:59 AM »

Good point! I just checked with a digital multimeter and it shows my AC line voltage is 126V (!). Will recalibrate the VTVM for that. The schematic is readily available, just go to http://bama.edebris.com/manuals/, then type in "Eico 232" and you'll find it at the back of the user's manual, if you're interested. It's a nice little meter, is proving accurate on all ranges (DC, AC, Ohms) and very stable after a few minutes of warmup.

73 DE Martin, KB1WSY
Logged
G3RZP
Member

Posts: 4715




Ignore
« Reply #36 on: April 25, 2012, 02:20:40 AM »

Martin,

For some reason with this new machine, I can't get .djvu files to open, so I can't look at the schematic. If the case is connected to the common input lead, that is a problem, and you pretty well have to stick to the 2 wire cord - which is not nice. However, the case should not be connected to the input return, and in that situation, a three wire cord with a ground has safety advantages.

All the VTVMs I have worked on have had the circuitry 'floating' with respect to the case, and teh case well grounded. If the connection is only at one point, it may be possible to float the whole circuitry away from the case.
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #37 on: April 25, 2012, 04:14:52 AM »

Peter,

I've emailed you the circuit diagram off-forum. The main probe lead is a shielded coax with the shield connected to the case via a coaxial connector (in my case, I replaced the original "solder dot" connector with a modern BNC) but with an open end i.e. the coax shield goes as far as the Eico "Uniprobe" (which has a switchable 1 meg resistor) to prevent stray currents, but ends there. The other lead is connected directly to the case with a banana plug.

My current feeling is that, if this unit was "safe enough" for 1950s technicians who were taking proper precautions, then it is "safe enough" for me but it is still an interesting insight into the kinds of risks that were considered acceptable "back in the day"! Still, it would be interesting to see if it can be brought up to modern safety standards.

Martin, KB1WSY
Logged
G3RZP
Member

Posts: 4715




Ignore
« Reply #38 on: April 25, 2012, 05:57:05 AM »

Martin,

emailed a reply to you. A 1500 volt range when using phenolic switches is not, in my opinion, safe. Nor is a floating case.

The weak point with all these things is what happens when you get an AC supply to chassis short - e.g a breakdown on the phenolic power switch wafer, or transformer insulation. Over here, anything with an ungrounded metal case has to have 'double insulation' to prevent that. Admittedly, 120 is not quite as bad as 240, but can still kill.

Good luck
73
Peter
Logged
K8AXW
Member

Posts: 3900




Ignore
« Reply #39 on: April 25, 2012, 09:04:11 AM »

Martin:  I too had a problem with the DJVU extension.  However, I checked the schematics of the other two Eico 232 and both used a power transformer for AC to the rectifier.  You might want to consider getting a similar transformer and making such a modification.

I've encountered this "hot chassis" thingy way back when and it's an unpleasant surprise!  Like RZP says, (paraphrased) 120VAC can still kill you but in my case the most unpleasant outcome was throwing the tool in my hand. 

Since then I stay away from things like this.  If you research radio and electronics to it's infancy, you'll find they used to do some pretty hazardous things and consider it the "norm."
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #40 on: April 26, 2012, 04:47:14 AM »

Allen:

>>both used a power transformer for AC to the rectifier.  You might want to consider getting a similar transformer and making such a modification.<<

Assuming that you mean I should add a 1:1 isolation transformer ahead of the twin diode that rectifies the measured AC, that sounds like a solution that would enable me to ground the case -- a bit of a challenge to find enough room for the transformer on the chassis, but just doable, I think. But wouldn't that also require the addition of a dedicated AC probe socket on the front panel? (Currently everything goes through a single socket connected to the nice, switchable DC/AC Eico "Uniprobe.) Or can I just add the transformer, then connect one side of the secondary to ground and the other side to the probe? While still being able to ground the case? Presumably that is exactly the point of using an isolation Xfmr?

I appreciate the responses I have had concerning this issue. I would prefer not to go SK prematurely, and have heard quite a few stories about "interesting" personal encounters with HV (or even just the plain old house AC supply) from old-timers -- enough to give me pause!

73s DE Martin, KB1WSY

Logged
G3RZP
Member

Posts: 4715




Ignore
« Reply #41 on: April 26, 2012, 06:27:17 AM »

Martin,

The VTVM ciruitry is isolated from the AC line by the transformer, so there isn't a requirement for another one. But for calibration, get a transformer with an output between say 30 and 120 volts. Measure the secondary voltage with the DVM, and also the VTVM and then adjust the AC CAL pot to get the VTVM calibration right.

Sounds as if you're having a ball learning new skills - keep it up! It's what ham radio is supposed to be about...
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #42 on: April 26, 2012, 01:52:08 PM »

So ... vanity is getting the better of me and I have posted some pics of my recent EICO test equipment renovation projects including the aforesaid VTVM:

https://picasaweb.google.com/114229926311767860862/WorkshopPix?authuser=0&authkey=Gv1sRgCN_up_fW4N32RA&feat=directlink

Unfortunately I am having so much fun with this that it will delay the famous goal of "getting on the air." I figured I needed test equipment before homebrewing my radio stuff .... but went a bit overboard. When I finally have my little RX and TX built, they will have been "tested" in every which way, before I even issue my first CQ!

73s DE Martin, KB1WSY
Logged
KB1WSY
Member

Posts: 804




Ignore
« Reply #43 on: May 17, 2012, 03:30:50 PM »

Does anyone know a source for male BNC plugs that would have two solder (or screw) lugs inside the barrel rather than being the "push and compress" type? I have made several attempts to install standard BNC "push and compress" connectors on coax cable and they have all failed: the pressure from the compressor part of the plug ends up shorting the center wire with the shield. Probably I am just inept about this, although the connectors came with instructions and I followed the instructions to the letter. I would much prefer something where I can see what I am doing. (This is partly because I want to replace the old "solder dot" connector on an old VTVM high-voltage probe and I want to be in no doubt that the connections are viable!!)

I've searched the Internet for BNC plugs that have "visible" screw or solder connections and have only been able to determine that they do indeed exist. What is eluding me, so far, is a supplier for these.

Thanks.

73 de Martin, KB1WSY
Logged
Pages: Prev 1 2 [3]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!