Heathkit IM11 vtvm filter question

[tubefan]

I think it would be ok maybe better filtering but want to ask the experts before I do. Heathkit im11 vtvm has an original cardboard 16uf @150v electrolytic in the power supply. Can i use a 22uf or better? Any thoughts are appreciated.

Thanks

[Sandbagger]

Jun 8, 2019 17:39:17 GMT -5 tubefan said:

I think it would be ok maybe better filtering but want to ask the experts before I do. Heathkit im11 vtvm has an original cardboard 16uf @150v electrolytic in the power supply. Can i use a 22uf or better? Any thoughts are appreciated.

Thanks

View Attachment

Filter cap values are not critical. You can substitute a higher value and you will get even better filtering. The only potential downside comes if you use a value that's a LOT higher, you can filter so well, that the voltage may rise higher than it should be. But going from 15 to 22 uF should not cause that problem. Make sure the voltage rating is at least as high as the original part. Higher is better.

[2600]

If this meter is as old as I think it is, the rectifier D1 will be a selenium type. That type rectifier usually takes the form of a stack of small parallel metal plates.

These will exhibit a large forward-voltage drop, typically 20 Volts or more.

If you had a "perfect" rectifier, one that does not reduce the peak AC voltage from the transformer, your 120-Volt secondary winding would put over 160 Volts DC onto that 150-Volt filter cap. In this meter, that 20 or 30-Volt drop inside the selenium rectifier is subtracted from that DC voltage, and the filter only has to tolerate roughly 130 Volts DC. Plenty safe for a 150-Volt part.

Making the capacitance value larger only serves to reduce the ripple voltage on the circuit in the meter. Since the meter circuit draws very little current, the seemingly-small 16 uf rating of the filter does the job just fine. Making the capacitance value higher won't make a difference you can see in any way. Won't hurt, up to a point.

A larger value will put a little more stress on the rectifier. When the power is first applied, the filter capacitor has no stored charge, it's effectively "empty". When the power is first applied, it must charge up from zero to full voltage. This causes a brief surge in current drain on the rectifier. A bigger filter makes this surge current higher. But making the value of the filter a little higher, like 22 or even 47uf should not create a problem.

It's common practice to replace a bad selenium rectifier stack with a more-modern silicon rectifier diode. Especially if it goes bad.

No need to fix something that ain't broke. But when you do this, you also eliminate that forward-voltage drop. A silicon rectifier won't lose more than about 1 Volt, usually less. The filter capacitor will now have a higher DC voltage on it than before. In this case, if D1 went bad and you put a silicon-diode rectifier in its place, the DC voltage on the filter would now be more than the rated 150 Volts.

In that case, the next-higher voltage rating would be wise when replacing C1. For that matter, even if D1 is okay today, it may decide to retire some time later. A replacement filter with a rating of 200 Volts or higher would still be okay to leave in the circuit if you changed D1 to a more-modern silicon part later on down the line.

If C1 has been changed some time in the last 10 years, there's no good reason to change it now. And if it is the original one, with a cardboard outer sleeve, look for a four-digit number on the markings. Typical pattern is two digits for the year, followed by two digits for the week of that year, 01 to 52. Sometimes it's seven digits. The first three identify the part's manufacturer, followed by the four-digit date code. The only three-digit manufacturer ID I remember off the top of my head is "235", which would identify Mallory as the part's maker.

So, if you see "235-6803" just as an example, it would be a Mallory cap made the third week of 1968.

That's a cool old meter. If that filter is original, go ahead and change it. You have lots of leeway as to what can substitute for the original part.

73

[tubefan]

Thank you for the help. Will change out the original cardboard cap. According to the parts list in the manual the diode is a silicone. Will look for a date code on the cap when I change it out.

[2600]

Ahhh, I would be skeptical of a 150-Volt rating with a silicon-diode rectifier. Makes me think that either the original part exceeded the marked rating for years, or maybe the 120-Volt power-transformer voltage is lower in the version you have.

Don't know. The math says that filter should have over 160 Volts DC on it. A 150-Volt part would not by my first choice to replace the original.

73

[tubefan]

Well I changed the cap. I ended up using 2 10mf@450v in parallel.  And you were right. It is a selenium rectifier.  So if I wanted to change the rectifier to a newer one what would you recommend?


Also a couple of quick questions. I know this meter is not a 2019 high end calibrated piece of test equipment but being a nut I want to get it as calibrated as I can. Using the method in the manual for DC volt it tells you to use a brand new 1.5v battery on the 1.5 volt scale and set the internal DC cal to a red dot on the meter. Once I do this the 1.5 scale is accurate but when I flip to the 15v or 50v scale using a new 9v battery I am about 1.5v low on the meter. There is not an adjustment for each scale so my mind is thinking maybe a high carbon comp? Or maybe I am just nuts wanting this meter to read close.


One other thing I notice is if I move my hand in front of the meter even powered off my hand can make the movement go up a bit on the scale and it will stay there. Like a magnet. I need to move my hand around until the meter sets back to normal and then keep away from getting too close. Is this normal for a vintage unit like this?


Thanks again


Attachment Deleted

[tubefan]

I guess this answers my question about the meter.

[2600]

I use a small pc board that has a 1/10 of a percent accurate reference-voltage chip on it. The guy who made these is no longer in business.

Had a quick look at Fleabay. Figured a chineseum equivalent might show up there. Didn't find one first try.

Hate to say it, but a 20-dollar digital multimeter is typically accurate to a percent or so. The accuracy is the cheapest part of a digital meter. What makes the expensive ones different is mechanical robustness and surge protection.

A cheap digital meter from Horrible Fright or Ebay might be the cheapest calibration solution. I wouldn't trust a flashlight cell. They tend to show more than 1.5 Volts when they're fresh and unused.

73