Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Dec Mon 09, 2019 5:21 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 40 posts ]  Go to page Previous  1, 2
Author Message
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Oct Wed 30, 2013 1:42 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 30698
Location: Pocasset, Cape Cod, MA
After looking at the schematic, I'd guess this thing puts out a negative 1kV pulse about 50 mS long, and I don't see any way of generating a higher voltage than that. The capacitor under test is going to discharge into that 500 ohms pretty fast, depending on its value.

The meter response would seem to depend strongly on the characteristics of those big capacitors; at the time this instrument was built, those were not the most stable or reliable components around.

It's an interesting exercise, but I think Johnny has the right idea.


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running
PostPosted: Oct Wed 30, 2013 3:49 am 
Member
User avatar

Joined: Apr Sat 17, 2010 10:56 pm
Posts: 192
Location: Paragould, AR 72450
Willitwork wrote:
Can you help us have some luck? Did you find something everyone should be aware of to avoid the learning experience?

I just happened to not be facing my workbench. The 383A was sitting on a small table next to my chair. I had just got it running again (just some bad solder joints). Later, I was preparing to take voltage measurements and when I grabbed the thing to move it onto my bench my hands ended up where they shouldn't have been. I stupidly forgot that I still had the thing plugged in and turned on and didn't check before trying to move it. :oops: Anyway since the thing wasn't sitting on my benchtop when I grabbed it, I was able to drop it almost instantly. Man, I still remember what it felt like to have all that juice running through my body... Getting 30kV from a weak CRT anode lead was nothing next to this. I honestly thought I was going to die. If that demon box had been sitting on the bench and me in my chair, I wouldn't have been able to drop it and I'd have sat there and roasted.

/lifestory

_________________
-Jim


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Oct Wed 30, 2013 10:17 am 
Member
User avatar

Joined: Jan Thu 24, 2013 4:55 am
Posts: 1338
Location: South New Jersey 08077
7jp4-guy;
Quote:
Out of curiosity, what make / model was the DMM that you managed to fry? Hopefully not a nice meter like a Fluke...

No, not a Fluke. It is/was a RSR-1007 from Electronix Express, made in Taiwan, it says. Inside on the pc board it is a "Model: 20". It is several years old and I got it mainly for the low ranges of resistance, capacitance, and inductance, but I have a better meter for components now.
Why didn't I use a HF Cen-Tech? Next time I will, it also claims to be good to 1kv.

The Pulser is not the only way this thing can bite the hand that fixes it, it also puts 300 VAC from the power transformer on the test terminals for measuring capacitors on the 1000mmfd and below scale.
And the rectified version of that 300 is used for the Megohm leakage measurement.

I measure 95 uHy on that output R6, 600 ohm 5 watt, I would also guess wire wound.
So if that rings when the pulse dumps through it, then the peak catcher diode I had on it could have been accumulating elevated voltage on the external 0.1 mfd which is where the DVM was hooked and the arc flashed.

My other hi-pot megohmmeter only does 500VDC and generates it with AA cells and transistor(s), It has never been as dramatic as this Simpson 383-A.
I have also seen a picture of the Simpson 383, which only did the Pulse test, I don't think I want one of those.
But maybe a six foot Tesla coil for RFI testing would be fun............

_________________
Ex: Same job, same building, same phone #; 4 Corp's: RCA, GE, Martin, Lockheed-Martin: 48+yrs '14


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Oct Wed 30, 2013 3:39 pm 
Member

Joined: Mar Tue 18, 2008 12:15 pm
Posts: 170
Wow. I read this with interest because I have a 383A that is on my bench right now, and I was just going to work on calibrating it.
I think that maybe I'll case it up and go back to my Sprague TO-6 or Hewlett Packard 4260A bridge.
Anybody interested in buying it, (the Simpson 383A) shoot me an offer!


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Wed 27, 2013 10:00 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4236
Location: Littleton, MA
Williwork's post, plus the fact that I'd never played with a thyratron before, stimulated me to restore the Capacohmeter I had on my shelf.

I got the schematic and a manual from BAMA. The Simpson 260 site also has a scan of the manual, but the one on BAMA is higher-quality.

I think the odd "pulse test" is just Simpson's marketing department making a virtue of a cost-cutting necessity. I suspect Simpson came up with this clever thyratron relaxation oscillator circuit to provide an inexpensive way of regulating the voltage applied to the capacitor under test, rather than pay for an electronically-regulated HV supply, an unregulated Variac-driven HV supply, or even a HV supply with enough current capacity to drive a wire-wound potentiometer voltage divider.

Image
The unit was pretty clean. I chopped off the old two-wire line cord and later replaced it with a three-wire cord. I had to run the ground wire half-way across the chassis to get to a convenient screw for the ground lug.

Image
There are some high voltage capacitors in the unit that would be a bit hard to replace, like the 0.1 µF 1000 VDC cap in the lower left of this photo. I didn't have a kilovolt supply to fully test the leakage on these, but at a couple of hundred volts they don't show any leakage.

I reformed the low voltage and 300 volt electrolytic capacitors; they came up fine. Simpson used good quality components.

The resistors were all still within the specified tolerance limits.

Image
Even the black beauty cap at the upper left of the photo showed no leakage.

I replaced the low-voltage selenium rectifier at the lower center of the photo with a 1N4003. I didn't bother with a series resistor, though I should have.

Image
I replaced the 300 volt selenium rectifier with a 1N4007, again without a series resistor.

The unit calibrated fine for the capacitance and megohm meter functions.

I had advised Willitwork to try using a peak detector and a voltmeter to measure the peak voltage, since the repetition rate of the pulses is quite low, about 0.2 Hz when turned up to the max. of 900 volts. But when I looked to do this myself, a quick calculation convinced me that since I don't have an electrometer (with its very high input impedance) I didn't have a chance of making a peak detector that would hold a voltage long enough for me to get a good reading. So, I had to use my scope.

For the high voltage pulse test, I didn't want to trust my cheap import 100:1 scope probe, as it's only rated for 600 volts max.

Image
So, I cons'd up a 100:1 divider using a Caddock 5 megohm 1% HV resistor that I bought as surplus, a 47.5K 1% metal film, and a Bournes 5K 10-turn trim-pot. I trimmed the pot to get exactly 50.5K ohms in the bottom arm of the voltage divider for a 99:1 ratio with the 5.00M ohm resistor.

I checked the divider using my scope's calibrator; it didn't seem to need a compensation capacitor to show a good square-wave response.

Image
I used this divider and my scope to measure the peak voltage of the pulse test. The calibration was still spot-on.

Regardless of the merits of the pulse test, I think the megohm function of the Capacohmeter is useful enough to earn it a place on my workbench.

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Thu 28, 2013 12:04 am 
Member
User avatar

Joined: Jan Thu 24, 2013 4:55 am
Posts: 1338
Location: South New Jersey 08077
Nice work on the divider, I was thinking that I would need to do something like that if I went back to running the "Demon Box" as one ARF'er called it.

I have discovered one other thing about the explosive arc event I reported in this thread (which kind of got me to back off of working on my 383-A):
A much closer look at the big 0.1 ufd 1Kv ceramic disc cap I tried for the peak detector showed me a hairline crack in the edge, opposite side from the lead wires.
I now believe that this crack is where the arc flashed out. This assumed INTERNAL failure of the cap would mean that I was wrong when I thought the arc jumped the gap between the leads.
So the good news is that we probably do not have to worry about how something like 30KV got generated, only something over 1kv.
But the control was set for the bottom end of the range, about 100V. Maybe a problem there.

Then there is the question; how good was my cap? They are marked .1Z Z5V MEXICO and with a very small logo.
The logo is like two isosceles triangles pointed to the left and right with the bases joined in the middle. The right side is filled-in dark, the left is open. the top two line segments are heavy, like a little roof over the two triangles.
I hand-marked them as 1KV when I bought them at the computer/ham festival several years ago.

I do however think if a peak detector is used again, that there should be a capacitor-under-test in place for a 'normal' load on the 383, with the peak detector across that.

_________________
Ex: Same job, same building, same phone #; 4 Corp's: RCA, GE, Martin, Lockheed-Martin: 48+yrs '14


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Thu 28, 2013 1:30 am 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4236
Location: Littleton, MA
Willitwork wrote:
Nice work on the divider, I was thinking that I would need to do something like that if I went back to running the "Demon Box" as one ARF'er called it.

Thanks! I'd mail it down to you if you want to borrow it. Let me know.

I intend to try to reproduce your experiment with the peak detector (except the part about blowing up the multimeter) once I figure out how to instrument it safely.

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running
PostPosted: Nov Fri 29, 2013 12:12 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 6373
Location: Wilmington, NC 28412 USA
anhero wrote:
Willitwork wrote:
Can you help us have some luck? Did you find something everyone should be aware of to avoid the learning experience?

I just happened to not be facing my workbench. The 383A was sitting on a small table next to my chair. I had just got it running again (just some bad solder joints). Later, I was preparing to take voltage measurements and when I grabbed the thing to move it onto my bench my hands ended up where they shouldn't have been. I stupidly forgot that I still had the thing plugged in and turned on and didn't check before trying to move it. :oops: Anyway since the thing wasn't sitting on my benchtop when I grabbed it, I was able to drop it almost instantly. Man, I still remember what it felt like to have all that juice running through my body... Getting 30kV from a weak CRT anode lead was nothing next to this. I honestly thought I was going to die. If that demon box had been sitting on the bench and me in my chair, I wouldn't have been able to drop it and I'd have sat there and roasted.

/lifestory


Do Not Think the 383A is to blame for your Shocking experience.

Bob T

The 383A I have behaves very well. I admit I use other Cap Meters on the bench, I have used this one without any problems.

Must admit the probes are lethal if you happen to contact them while doing Pulse test. As many early ecectronics were in that era.

Not for the inexperienced DYI for sure.


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Fri 29, 2013 12:45 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 6373
Location: Wilmington, NC 28412 USA
While PRICE is CRAZY the Write up Discription is very interesting.

http://www.ebay.com/itm/Simpson-HIGH-VO ... true&rt=nc


Bob T


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Sat 30, 2013 2:27 am 
Member
User avatar

Joined: Jan Thu 24, 2013 4:55 am
Posts: 1338
Location: South New Jersey 08077
I was looking carefully at the pictures Steve Byan posted and looking at the lower left corner of this one:

Attachment:
img_0743_med.jpg
img_0743_med.jpg [ 61.66 KiB | Viewed 3313 times ]


I see the two 510K resistors on the terminal strip (R3 & R4 in the 383A)
These look to be original, and look exactly like the ones in my un-molested parts-donor 383A, and like other photos I have seen.
In my parts-donor, they are 0.4 inch long, and look like carbon comp types to me and are just listed as 1/2w, 10% in the parts list.

In the capacohmeter that I have been working on, the R3-R4 pair of resistors is now replaced by three series resistors, 1/2w, carbon comp NOS which are 220K + 330K + 470K. This choice may be sub-standard, as has been pointed out in earlier replies, if the full B+ were across the string.

I had assumed that the original R3 & R4 were 'ordinary' carbon comp, and so thought using three in series gave me a safety margin compared to the two original, for the voltage rating (and power rating too). If the original 510K parts were rated for 500V then the string could take a current of 0.98mA and not be over-voltage. The current should NOT be continuous, it would put 0.51 w in each resistor, but this is a low duty factor pulser.
That current would put 460v across my 470K resistor, so the other half of the question is what is its voltage rating?
We don't know. As others have posted, it sounds like 500 and 350 are/were common voltage specs for carbon comp resistors, from various manufacturers and various eras. I suppose that to make the safer choice, I should assume the old resistors were used in the 383A at up to 500 v and that the slightly different style, probably newer, resistors I had on hand should be used at no more than 350v. I could split the string up differently (2x220K + 2x330K=1100K), or rob the parts-donor, or just order 510K 1/2w, 500v (or 1w) new parts.

What has anyone experienced as far as series over-voltage effects (not break-down to adjacent circuits) on carbon composition resistors?
They could breakdown without being over-temperature, because in a pulse circuit the dissipation wouldn't last long enough to over-heat them.
Do they split open? do they just go up/down in value? Or do they arc across (internal/external?), but return to normal after the arc?
What should I look for as possible symptoms of over voltage exposure? External carbon tracks?

_________________
Ex: Same job, same building, same phone #; 4 Corp's: RCA, GE, Martin, Lockheed-Martin: 48+yrs '14


Top
 Profile  
 
 Post subject: HF givaway meter at a grand
PostPosted: Nov Sat 30, 2013 3:09 am 
Member

Joined: Mar Sat 06, 2010 11:34 pm
Posts: 282
Location: Las Vegas, NV
HF givaway meter at a grand

Attachment:
grand1.png
grand1.png [ 161.41 KiB | Viewed 3309 times ]


HF giveaway meter at 1000 VDC The source is a Fluke 332B and should be right give or take 20mV or so

BTW: The modern (last 30 years or so ago) Flukes have a 1K fuseable resistor that protects the rest of the meter from massive overloads. Change it and your good...


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Sat 30, 2013 4:11 am 
Member
User avatar

Joined: Jan Thu 24, 2013 4:55 am
Posts: 1338
Location: South New Jersey 08077
Not too bad a percentage of error on that HF meter, in terms of $/V.

Now I remember why I wasn't using the HF DVM when I blew up the capacitor attempting to measure the pulse peak voltage, the 1 meg-ohm input resistance.

_________________
Ex: Same job, same building, same phone #; 4 Corp's: RCA, GE, Martin, Lockheed-Martin: 48+yrs '14


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Nov Sat 30, 2013 4:20 am 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4236
Location: Littleton, MA
Willitwork wrote:
What has anyone experienced as far as series over-voltage effects (not break-down to adjacent circuits) on carbon composition resistors?

Horowitz and Hills "Art of Electronics" has a plot of current versus voltage (IIRC; it might be resistance vs voltage) for carbon composition resistors. Their resistance isn't constant as the applied voltage increases; the resistance of a carbon composition resistor decreases as the voltage across it increases above a threshold.

Here's another reference with some info:
http://www.barthelectronics.com/pdf_fil ... 20Page.pdf

So, over-volting the resistors might not display any damage, but may result in them having an effective resistance that is significantly lower than their resistance at the spec'd max voltage.

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Sun 06, 2014 10:10 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4236
Location: Littleton, MA
Update:
The pulse test showed some leakage even on a good capacitor, so my attempt to reform the 2000 µF 6 wvdc electrolytic capacitors didn't pan out.

Image
I replaced them with modern 2200 µF 10 volt capacitors. Compare the size of the new ones to the old capacitors shown in the earlier photos above.

Image
I didn't have any 22 µF 25 v capacitors on hand, so I left the originals in the firing circuit for the thyratron. I did, as a precaution, replace the Black Beauty filter capacitor on the low-current 300 VDC supply for the megger function. However, once replaced, I couldn't zero the resistance scale! While trying to diagnose the problem, I discovered that putting the 10 megohm input resistance of my Fluke 27/FM meter across the 300 volt supply allowed me to properly zero the scale. Apparently Simpson designed-in the leakage of the Black Beauty paper cap! So, I added a 10 MΩ resistor across the new polyester film filter cap to simulate the leakage of the old capacitor.

Measuring it out of circuit, the Black Beauty only showed 0.4 µA of leakage at 300 volts. That's about 7.5 MΩ.

Image
The cap is actually a Pyramid "IMP" rather than a real Sprague Black Beauty, which someone on the internet claims is a di-film (mylar and paper) capacitor. If that's correct, I'm surprised that the leakage is that high.

Pulse mode experiments

Willitwork had been trying to calibrate his 383A using a method I suggested - a peak detector consisting of a diode, a capacitor, and a DMM. (At the time I suggested this, I didn't know that the 383A has such a low pulse rate at high voltage; the time constant of a reasonably-sized cap with the 10 MΩ input impedance of the DMM is just too short for this method to be effective.) He reported some surprising results - large loud sparks and a blown-up DMM :-(

Later, I read this caution on page 34 of the manual: "Do not apply the Capacohmeter leads across a transformer winding, inductance, or rectifier circuit. The output from the instrument can produce undesirable results for such connections." The caution about inductances I can understand, but I don't get the caution about rectifiers.

I wanted to understand what was going on, as I couldn't see anything in the circuit that would multiply the 383A pulse output voltage to the point where a spark would jump a large gap. But I also didn't want to blow up my meters or scope. My homebrew high voltage 100:1 divider provided the solution. In the meantime I had also obtained an HP 1741A storage scope, which is perfect for examining a low-repetition-rate fast pulse signal.

Image
I wired the divider across 0.1 µF 2 KV ceramic cap, and a 1N4007 diode in series with them.

Image
The divider feeds channel 1 of the HP 1741A scope.

Image
I started by shorting the 383A output and zeroing the meter on the megohms function. This adjusts the internal voltages to compensate for variations in the power line voltage.

Image
Next, I connected the 383A across the capacitor and paralleled divider. The megohmmeter properly reads 5.05 megohms.

Image
Setting the 383A to 500 volts "PULSE", the meter shows the cap to be OK. I guess 5 megohms leakage isn't very much, and the leakage is symmetrical, so it balances out of the pulse test circuit.

Image
Here's the pulse waveform, at 100 volts/division vertical and 1 ms/division horizontal. The 5.05 MΩ impedance of the divider knocks the peak pulse voltage down to 450 volts. When I measure it with a 10 M 100x scope probe, it's spot-on at 500 volts. You can't see it at this scale, but speeding up the horizontal sweep reveals a very small ripple at about 100 kHz on the decay. I suppose it's the result of a parasitic inductance resonating with the capacitor under test.

There's also some 60 Hz ripple from the odd circuit return connection to the thyratron filament supply rather than to ground. Maybe that's to synchronize the firing of the thyratron to the 60 Hz power line frequency, so that old service grade scopes could use a line-triggered sweep when calibrating the pulse voltage?

Image
Now for the moment of truth. I connected the 383A to to diode and turned on the pulse.

Image
This time, the asymmetrical leakage due to the diode shows up as a "bad" capacitor.

Image
At the same 100 volts/division vertical and 1 ms/division horizontal settings, you can see the peak capacitor voltage is still 450 volts, but the decay is much longer due to the diode isolating the capacitor from the 383A discharge path.

No fireworks, but I didn't turn it up to the full 900 volts. Still, I don't see anything that would lead me to think there would be any problem at 900 volts.

Maybe I should try the experiment with the diode reversed - I wonder which way Williwork had his connected?

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Last edited by stevebyan on Jul Wed 09, 2014 1:44 am, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Mon 07, 2014 1:09 am 
Member
User avatar

Joined: Jan Thu 24, 2013 4:55 am
Posts: 1338
Location: South New Jersey 08077
Interesting results, certainly appear reasonable.

I'll add a reminder that I revised the arc description:
Quote:
A much closer look at the big 0.1 ufd 1Kv ceramic disc cap I tried for the peak detector showed me a hairline crack in the edge, opposite side from the lead wires.
I now believe that this crack is where the arc flashed out. This assumed INTERNAL failure of the cap would mean that I was wrong when I thought the arc jumped the gap between the leads.
So the good news is that we probably do not have to worry about how something like 30KV got generated, only something over 1kv.
But the control was set for the bottom end of the range, about 100V. Maybe a problem there.

Then there is the question; how good was my cap? They are marked .1Z Z5V MEXICO and with a very small logo.
The logo is like two isosceles triangles pointed to the left and right with the bases joined in the middle. The right side is filled-in dark, the left is open. the top two line segments are heavy, like a little roof over the two triangles.
I hand-marked them as 1KV when I bought them at the computer/ham festival several years ago.

I do however think if a peak detector is used again, that there should be a capacitor-under-test in place for a 'normal' load on the 383, with the peak detector across that.


As for the peak detector diode polarity, it was all a clip-lead lash-up, so no way to get absolute proof now, but I recall being aware of the negative direction of the pulse. That makes me think the pulse output of the 383-A went to the cathode of the 1n4007.

I have been wondering if failure of the thryatron to trigger could be involved, if it skipped a pulse, then the output voltage would charge higher than indicated by the pulse control. Eventually, it would fire, and the output would be higher than what was set on the pulse voltage dial. It may not need anything more exotic than a faulty
pot to give excessive pulse voltage. And my "1KV" cap did not have much margin if the 383A was not actually dialed down to a few hundred volts.
A GE GL-2050 tube spec was easiest to understand; showing more negative grid voltage to hold-off the firing of the tube, till higher positive plate voltage.
But the RCA spec file is smaller.

Read the first graph as showing the "CRITICAL" grid voltage that will hold-off the firing, even up to the anode voltage on the Y axis.
Attachment:
2050_RCA_-00173562.pdf [222.04 KiB]
Downloaded 46 times


Notice I have NOT removed the "CAUTION" from the title, yet. :shock:

_________________
Ex: Same job, same building, same phone #; 4 Corp's: RCA, GE, Martin, Lockheed-Martin: 48+yrs '14


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Mon 07, 2014 9:27 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 30698
Location: Pocasset, Cape Cod, MA
I don't know if "rectifier" meant "silicon diode" but if it did, those were made differently when this instrument was new. If you exceeded the PRV, the junction could short. Modern diodes go into avalanche when their PRV is exceeded, which doesn't hurt them as long as the power dissipation is low enough.


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Mon 07, 2014 9:54 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4236
Location: Littleton, MA
Willitwork wrote:
II have been wondering if failure of the thryatron to trigger could be involved, if it skipped a pulse, then the output voltage would charge higher than indicated by the pulse control. Eventually, it would fire, and the output would be higher than what was set on the pulse voltage dial. It may not need anything more exotic than a faulty pot to give excessive pulse voltage. And my "1KV" cap did not have much margin if the 383A was not actually dialed down to a few hundred volts.

The voltage pot on my 383A has 15 or 20 degrees additional rotation past the "900" volt marking, but if I turn it up past 900, the thyratron never fires. I guess that means the HV supply can't provide much more than 900 volts, so when the thyratron threshold voltage is turned up, the HV supply never charges the cap high enough to fire the tube.

But your 383A might be different; I didn't replace the HV caps, so my unit might have enough capacitor leakage to keep the peak voltage down to a little over 900 volts.

Yeah, it does sound like you had a bad cap, maybe overstressed a little on the voltage, too.

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Mon 07, 2014 9:57 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4236
Location: Littleton, MA
Alan Douglas wrote:
I don't know if "rectifier" meant "silicon diode" but if it did, those were made differently when this instrument was new. If you exceeded the PRV, the junction could short. Modern diodes go into avalanche when their PRV is exceeded, which doesn't hurt them as long as the power dissipation is low enough.

Good point.

What does exceeding the P.R.V. do to a selenium rectifier?

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Mon 07, 2014 10:42 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 530
Location: Monte Vista, CO. USA
I think it shorts and releases toxic smoke, which is something selenium rectifiers can do even if they are not being stressed.


Top
 Profile  
 
 Post subject: Re: Simpson Model 383-A Capacohmeter running - CAUTION
PostPosted: Jul Tue 08, 2014 2:39 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 30698
Location: Pocasset, Cape Cod, MA
I believe they're fairly tolerant of voltage spikes.


Top
 Profile  
 
Post New Topic Post Reply  [ 40 posts ]  Go to page Previous  1, 2

All times are UTC [ DST ]


Who is online

Users browsing this forum: bob91343, Peter Bertini and 10 guests



Search for:
Jump to:  
cron




























Privacy Policy :: Powered by phpBB