• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

866a valve test

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Not simple to test a mercury vapor rectifier. The best test is in a circuit. You need to apply filament voltage for a minute, then high voltage ac to see how it rectifies.

You can also assume the mercury is still there, and test for emission without ionizing. Compare results with known good tubes. Or bet on statistics and see how similar all your tubes are.
 
Iv tested 2 of the filaments so far with 2.5v ac and looks like there ok,there seems to be mercury in them still! how would i test for emission without ionizing? Is ionizing what it does when it turns bule inside when voltage is applied to the anode? I dont have any good way of testing them, was wondering whats the minimum voltage i can try on them? and how could i do it?
 
have to handle them with care ect how could i go about testing them?
Assuming you have no tube tester that will check them, here's a quick and dirty way of doing so. You'll need a 2½ volt transformer to power the filament. And you'll need a metered 100 volt AC (or DC) supply and a 100 watt wire wound resistor of at least 100 or 200 ohms. Light the filament and connect the tube across the 100 volt supply with the resistor in series. For 100 ohms you should draw almost 1 amp from the supply. With 200 ohms you'll draw ½ amp and so on. You could even use lower or higher voltage depending on what you have and/or feel comfortable with. But you'll want to draw at least ¼ amp through the tube for a meaningful test.

A word of caution. If you have modern day 866's with the smaller envelope, there's not much to worry about. But if you have new old stock from years back with the larger (fatter) glass envelope, these tubes have a problem. It seems that the coating on the plate as well as the cathode shield tend to flake and peel off. When this happens, the flakes can get between the plate element and the cathode and short the tube and your supply out. I've seen this happen many times with the older stock 866's. Even old 816's and 872's have this flaking problem. This may be hard to see at first because as the mercury vaporizes it clouds the inside of the tube. So you have to pay close attention if you have older stock. And even if the flakes go to the bottom of the inside, there's no guarantee they'll stay there. I would not use one in this condition.
 
Another comment - don't lay the tube on its side while testing. They should be upright.

If the tube shorts as HollowState cautions, you have no problem because the resistor will limit the current. It depends on whether you are using an ac or dc supply, just how much current you will measure. Since it only conducts on half the cycle of ac you will show a lower reading. Incidentally, a 100 Watt light bulb might be a good, cheap limiting resistor.
 
thats what i was hoping to use! just need to find 100volts now.

You can use light bulbs (the common incandescent domestic types of various wattages) in place of power resistor, they will work just as well on DC. Light bulbs are rated for tens or hundreds of watts of dissipation and don't cost a fortune in case they die.

Note that light bulb's resistance is considerably lower when cold, but it goes up as it warms up (PTC).

You should start with low wattage types first (these draw less current) and move on to higher wattages if everything works as expected.
 
Using an incandescent lamp as the series resistor, while it will function to a degree, will only give you empirical results that will have to be interpreted. Since the resistive element is auto-variable (spongy), current through the tube will vary as will as the voltage across it. A resistor of fixed known value will provide a much more discernable answer. So in other words, for a meaningful test setup there should only be one variable element, the tube itself.
 
Administrator
Joined 2004
Paid Member
Using an incandescent lamp as the series resistor, while it will function to a degree, will only give you empirical results that will have to be interpreted. Since the resistive element is auto-variable (spongy), current through the tube will vary as will as the voltage across it. A resistor of fixed known value will provide a much more discernable answer. So in other words, for a meaningful test setup there should only be one variable element, the tube itself.

I agree with Hollowstate - you absolutely do not want more than one variable in the equation if you are attempting to verify these tubes are good.

Also do not exceed the continuous maximum rated current as you test them.

I've had a few of those older NOS tubes with the flaking plates, sadly I'd pitch any encountered.

Be careful not to break them as the mercury is toxic. Safe diy clean up is possible see here for details on this and a lot of other useful things:
ClariSonus - Clear and enjoyable sound Power Sources

(Honestly speaking I have just a few good ones and I am going to get rid of them)
 
Last edited:
The tubes i have look used there not nos, Im just testing the filaments at the mo leving them on for 30 mins or so so the mercury is not stuck all arount the inside of the glass, I can seem to find any flakes! I have a 25w light bulb and a 40w and 60w and 100w and 150w and (300w (e40). Still need to find a transformer that will do 100v. I want to build a psu using them need 4 working ones to biuld another amp!!! I'l try to get some pics of them.

thanks
Dave.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.