B&K model 747 tube tester for EL34

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
For-get-it buy something that has either been tested on a metrix U61 funke w19 W20 neuremberger 370-375 or an AVo.

The reference might not make any sense. alot of this junk is stored for years and the deviation between two testers from the same lot from the same year might still be HUGE. dont forget the potentiometer used for setting the load can only be set reliably to +-5% and the POT is +-10% also.

Not very encouraging. I know. but there are alot of Fools on ebay

the EL34 is best tested.

at 250 volt anode and screen grid voltage and -13.5v G1

anode current should be no less than 50 mA or its ready for the dumpster. between 50 and 60 millampere it might be good if the mutual conductance is still strong.

from 60mA onwards it can be seen as good whit examples whit good lifetime should test strong on mutual conductance

Grid current may not be greater than 0.3uA @ -1,5v VG1


i think you should have them re-checked by somone whit a Rebuilt Hickock.or anything else from the good list.

V4lve.
 
For-get-it buy something that has either been tested on a metrix U61 funke w19 W20 neuremberger 370-375 or an AVo.

The reference might not make any sense. alot of this junk is stored for years and the deviation between two testers from the same lot from the same year might still be HUGE. dont forget the potentiometer used for setting the load can only be set reliably to +-5% and the POT is +-10% also.

Not very encouraging. I know. but there are alot of Fools on ebay

the EL34 is best tested.

at 250 volt anode and screen grid voltage and -13.5v G1

anode current should be no less than 50 mA or its ready for the dumpster. between 50 and 60 millampere it might be good if the mutual conductance is still strong.

from 60mA onwards it can be seen as good whit examples whit good lifetime should test strong on mutual conductance

Grid current may not be greater than 0.3uA @ -1,5v VG1


i think you should have them re-checked by somone whit a Rebuilt Hickock.or anything else from the good list.

V4lve.

I'm sorry, but the 747 is a very good tester. It is a mutual conductance tester, and tends to hold calibration well. It suffers some of the same shortcomings of many testers, but it's not "junk". As well, sometimes the tests that need to be done don't require sophisticated analysis.

The one annoying "feature" is the 747 and 747B use a meter that does not read or convert to absolute Gm readings. It has a BAD - ? - GOOD numeric scale. The minimum of the "GOOD" section is 65. There is no 100% per se, as the normal variance in Gm from tube to tube makes it hard to pick a specific number.

I have multiple Hickoks, an AVO CT-160, and a power tube test/match rig of my own design - in addition to a 747B B&K. All testers have their shortcomings - but you have apparently never used a 747 or 747B or you wouldn't make the statements about it that you did. For the kind of "go-no go" testing the original poster described it will do a FINE job. I find it to be a very useful tester.
 
point made.

However absolute Gm Readings wouldn't suit a piece of equipment made for general radio repair. be course people would represent these as absolute whilst they are not .


i must agree whit you for a go-no-go test they will do fine . but representing anything from these machines as absolutes is bad behavior :D


I'm currently also building my own valve tester. it is still under construction but i could spare some details and building ideas. PM me if you would like to swap some ideas :D I'd be happy to, becourse there arent that many Homebrew freaks around :D

You know how hard it is to make something that gives absolute results ?

For instance if i test Tube X whit my (very good) variable power supplies witch are 0,01% stable over short periods. and i AC heat the fillaments, switching on a hair dryer can mean the readings drop can drop by a few percent !

This is how hard it is. to give absolutes , Now here comes ebay seller Mister-x who gives his test results in 0.5% resolution :confused: From a Beat down tube caddy-toss tube tester . i find such things hard to accept :eek:



Moral of the story. dont even think about representing anything as absolutes if you have not checked all the operating characteristics against the (brand) specific data-sheet of your DUT.

V4lve.
 
Thanks so much V4lve.
I totally agree with your opinion and i am planning to build my own tube tester based on your idea.I found this useful example from a Diyer from China.
Let's share with all of you here.

µç×ӹܲâÊÔÒÇÖÆ×÷

That is one of the simplest and nicest vacuum tube tester i have ever seen. it even has short circuit protection ! :drink:

Im using a similair mosfet HV supply. but instead of the VB IC i used a 100K potentiometer set at 3 milliamperes of current through a LM317HV driving a mosfet source folower. whit exactly the same short circuit protection !

However a sidenote. i have Put a 100r resistor in The HV supply be course at 350 volt this will limit the peak current to 3.5 amperes (before the transistor latches on) and servos the current to the maximum (high voltage mosfet max peak current 5A)



see this PDF http://83.117.57.34/nvhr/TubeTester_rht136.pdf

Its a piece from the dutch radio history magazine. this man has build a tester where every voltage is DC and can be monitored for voltage and current!
he can test whit 700 volt 200 milliampere anode current !

I can maybye translate some of it .

v4lve
 
That looks very interesting if you could post a translation, at least some of it. I'm in the process of building something similar. Nice thing about this kind of "tester" is not so much testing as prototyping. Instead of simulating or building a stage, you can just "test it" and dial up the biasing for the actual tube you're going to use. Not for the theoretical curves in the data sheet.
 
Last edited:
I've been looking for this PDF, but it looks like the site went down :(

There was a translation posted at this site: Information on a Valve Tester please - UK Vintage Radio Repair and Restoration Discussion Forum

Nicko, here is a Google translated text from the Dutch original text, so there might be mistakes and missing or added words that are wrong:

----- Google Translate below -----

Hello friends tubes lovers

The idea to build yourself a tube tester is a bit out of necessity. I have for over 50 years a large collection of pipes together received most of the stock by buying from traders who stopped the case and sets on your left and right met. Myself have over 40 years had one thing in radio, TV and so-called white goods and are now more than 10 years hobbyist.

In the time I still in the repair Sat there was little need for a tester, when you repair a pipe in doubt took up a new one from the stock and the problem was solved or was elsewhere. Given my age, now 77, I thought it was time to get some of that inventory to sell. Since not all pipes, especially the older types, were not new had come to a tester the wheat from the chaff to separate.

I then bought a tester, an old army performance, the I-177. But nothing could, the tester is suitable for mainly American pipes of the time, and not really modern or old pen tubes. An extension set for that matter brought no result, no book!. A disadvantage, and that applies to many testers, is that you must have a book related institutions. And the tube you want to test must also occur in otherwise, you can do nothing. The same goes for testers with cards or rolls.

So the tester sold again and decided to build himself a tester. But only one where everything has to be tested. For setting, you only need the data from the pipe manufacturers and those in the various tubes guides beschikbaar.Voor than what you find there is not the Internet. It is only needed to put this data to the tester. It is best to do, but requires an extensive supply the voltages over a large area should be adjustable.

All transformers except the high voltage transformer and trafo'tjes to supply the digital meters from the demolition. The transformer for the power I have settled in Europe in AE Schagen. That give you what you want and they make for a very reasonable price. This transformer delivers 200 mA at voltages up to 700 volts. The small trafo'tjes have a high voltage between primary and secondary to be, and the entire measuring circuit should be very well insulated (at least 700 volts) are compared to the zero. For safety, I have every meter of its own power supply provided.

The transformer for the filament supply (T1) is a destructor from a large audio amplifier, the secondary voltages quite matched what I wanted. Moreover, there is a 2 Amp variac for, so is not as precise. Output of the rectifier is closed with a bleederweerstand of relatively low value, it is necessary because otherwise the capacitors back to controlling the tension slowly discharged.

The high voltage transformer (T2) is wrapped according to the specifications, with branches from 100 to 700 volts. The two capacitors are in series to two, because no 700-volt capacitors available. The resistance of the capacitors are required to differences in the internal resistance of the capacitors to eliminate. If the bleeder is on this tension regelpotmeter the g2 voltage (25 Kohm to 100 watts). What is generous, but that is my preference. That is a variac transformer of 2 pre-amps. Why all the variac's asking you might be wondering, there are still enough electronic circuitry to a DC rule. Right, but give me an old custom variac faithful, never breaks, and regulates grandiose is not averse to a bit of overload. The transformer parallel to the line side of the variac (T3) to operate a relay that prevents the closure test still can be operated when the power is on.

The transformer T4, also from the demolition, is a backwards from an old radio connected power supply with rectifier bridge. Were indeed two separate windings because I have the polarity of the voltage switchable g1 wanted to make. This transformer supplies 110 volts from the second winding, the power for the closure tester. This is not from a winding as a floating voltage to be compared to the zero. Transformer T5 provides the power for the relay of the closure test. Then the little transformer T6, there are six pieces, each one meter feeding. With the so-called tripod, the 7809, the voltage of each power supply stabilized at 9 volts. These things are unrefrigerated for a print, with this cooling load (approximately 65 mA) is not necessary. Then finally, T7, but that's not a transformer called SMPS, Switch Mode Power Suply, or switching power supply. This provides 5 Volts for the neat-gloeispannings or flow. The conversion of the meter of voltage to current occurred with a 4 pole switch that upon release returns to its original position. It is located left of the heater voltage meter on the front. It also shuts off the decimal point for current measurement shows a scolding, so read the voltage with 0.1 volt resolution (for the current 10 mA). For the heater voltage DC I have chosen the simple fact that with such a digital gauge is easier to measure. For a cathode remains the same no matter where you warm fires, with AC or DC, if necessary can do it with a gas burner. Battery tubes keep most of DC, AC on the filament has the same effect as AC on g1, so hum.

The power switch has three positions, in position 1, only the stress test for the closure site. This is negative, otherwise the tube will guide. In position 2, the heater voltage to power the meter on so you can test tube hot and cold on a possible closure between the electrodes. And in position 3 is there with the high voltage power supply for small meters. All devices are primarily fused.

It cost me about one year before I had all the parts together, building itself around one year. The biggest problem was the switch for connections to the tube feet, it must be of good quality in terms of isolation and switching capacity and mother also have contacts that break for it.

Preparing the front, the top plate for the tube sockets and chassis for food was easier said than done. First of all with a scribe drawing on the aluminum, and then cutting with a jigsaw, by filing tidy, and you have the rectangular holes for the meters. For the large round holes for the sockets I had an adjustable hole cutter and a large drill available, small holes such as the smaller tube feet and the capacitors, s are easy to drill, a ladder drill 5 to 35 mm are ideal. First I have begun a frame of 20 mm angle iron welded together, good angles and the right size, the plates and screws to a firm foundation for the whole. Of the chassis for providing power to the front and the top plate with the tube feet between the fully wired with plugs out so that they disassemble. Of this I have a later benefited greatly, because it was not so, that after installation and plug into the socket the unit also worked well immediately.

I had already taken account of the fact that when many sockets and connecting to it a bit neat to see the wiring still in knots together, you're creating an oscillator. If you build a circuit that must oscillate he does not, but if it can not happen spontaneously, it's all in the game. Had mounted several ferrite beads, but it proved not enough, some steep power tubes were spontaneously generated. To avoid influencing the measurements, I included no stopping resistance. But with a handful of beads, disconnections and by some extra lines to partition, I can solve this problem. I have the flow in the pipeline is set G1 had a lot of convenience. This test positive grid electricity and electricity is only one tube oscillates. No leakage so meter grid which, after I saw later on the forum, my whole issue was how this could occur. It came to me now in very handy, you could see right away if anything went wrong. If all goes well it stays at 0.00 mA assist the testing of a tube. Meanwhile I have the sensitivity of the grid where flow increased from 100 nA that can be read. Above 400 volts anode voltage you measure this mainly the leakage current of the tube feet.

The diagram of the tube you see where and how all meters included. I chose digital meters because they have much more accurate and can be read than analog meters.

In the input circuit of the test tube is also included with a 1:1 audio transformer at the input 2 BNC parallel buses. And at the anode via a coupling capacitor again 2 BNC sockets. The input you can connect an audio generator, possibly with a scoop, an interception and the output amplifier and the other input of the scope. So you can see exactly how much the tube reinforced. Instead of a scoop with multiple inputs can also use AC meters. Since all voltages are adjustable find fun to experiment and see which setting you get maximum gain from a tube or when the case is deformed by example overload or an incorrect setting.

Through the buses under the Ra EXT switch you can use a resistor bank to set any value for Ra. The output of a tube is shifted 180 degrees relative to the input signal. By inverting input of the scoop, you get the signals in phase to see. They can also move apart and then you see just a sine wave. This means that the amplified signal is undistorted. From the anode half goes to a capacitor integrated transistor amplifier for the control of cracking or microphonics' s tube. For this test to make sure that you turn on an anode resistor, otherwise you hear.

The voltage on the control grid in 4 steps of 25 volts to 100 volts to turn negative, positive or possibly to earth. (See diagram). Each step is regulated 25 volts. That tension, but then switched positive, I use to test the often sensitive detection diodes. Testing with normal Ia meter is what link, the flow is then not bounded.

Fine tuning of g1 is possible with a 3 turn potentiometer, which controls more closely. In the grid circuit are also two switches for the vacuum test set, NC (Normal Closed). This are two resistors, a high value (3m3) and one with a lower value (330 K) ohms. When the button is pressed on the 3m3 approach is therefore a strong resistance in the grid circuit are, a small change in anode current due to insufficient vacuum. Quite steep output tubes to give this value a bit too often, but this disappears when you press the button to 330 Kohm. This phenomenon is caused by the control grid in such tubes is very close to the cathode, so it is very hot and spontaneous is going to emit electrons.

This tester, I now have more than two years in operation and it suited him very well, really works great. I've also been a few people here in the hobbyhok had brought along their tester to compare, they doubted the results of their own meter.

With a little patience it is possible to complete an Ia-Ia or Va-Vg characteristic of a tube to take. It can then calculate the internal AC resistance.
Something about the front itself, looks like it comes straight from the factory. But this is made by an engraving. I had this or a drawing with exact dimensions to within 0.1 mm accuracy there. The buttons are called tool buttons, which have a central clamp mount. The advantage of these buttons is that they are absolutely stuck on the axle and not twist, some arrows with screw sometimes, especially if the switches a bit heavy going. The side panels are plastic plates 6 mm thick, the zgTrespa. Here I have a few big catches on screwed because the weight is almost 25 kg. On one side are a fan built in for cooling, only comes in when the power is turned on.

Would anyone want to build yourself a tester to suggest the same principle, and not a few gauges to cut. You will see at a glance all the correct values ​​of all voltages and currents as the manufacturer of the tube that entered, and then also very, very accurate. This I have seen in any factory tester. There are several things you could omit to simplify such short-circuit test, the audio input and output, and instead of many sockets fit only one foot and the rest gradients.

Jan van Eyck

----- Google Translate above -----
__________________
Martin, Sweden
 
There you go.

Recently I've started work on my own version of said tester.
I've so far completed the framework. dimensions are 50x35cm base 46cm high whit a 25x50 top plate.

I'm still in doubt whether I should have regulated screen and anode voltages as in the Metrix U61 or just variac controlled ones. Recently I've acquired two 0,7A variacs which are perfect for the job.

V4lve
 

Attachments

  • jan van eijck buizentester.pdf
    931.5 KB · Views: 87
  • IMG_0084.jpg
    IMG_0084.jpg
    564.3 KB · Views: 142
  • IMG_0085.jpg
    IMG_0085.jpg
    589.4 KB · Views: 140
The tester is nearing completion, and needs another 20 hours or so to get it done.

I've opted for a variac adjustable high voltage and filament supply.

So far, all the supplies are in working order, the unit features two valve regulated supplies capable of 90-320V one capable of 45mA the other 90mA (1 vs two 6v6 pass tubes)
and a variac adjustable supply of 0-500V 300mA CRC filtered.

Most of the knobs on the front are for range switching on the analog meters, and adjusting the supplies. There is a knob that acts as a shorts test, which individually connects every electrode to a high voltage potential through a neon bulb which will light up if the electrode is somehow shorted to the other elements.

The grid supply is a solid state shunt regulator for 0 to -85V I am unable to find schematics for it right now.

I have not drawn schematics for the major part of the tester because its merely a collection of power supplies, meters and a switch matrix to select the tube electrodes. The high voltage 0-500 volt supply is trivial to design if you want to build one of these boat-anchors.


The tube regulated power supply deserves some attention though, it consists of a classical tube regulator, enhanced by two constant current sources that increase the loop gain and line regulation of the regulator section. I have chosen E83F pentodes for the regulator tube mostly because of their stability over lifetime.

The -85V stabilized by a 85A2 (5681) is double stabilized against mains variations by means of the 0A2 tube.
 

Attachments

  • IMG_0970.jpg
    IMG_0970.jpg
    532.2 KB · Views: 109
  • IMG_0974.jpg
    IMG_0974.jpg
    564.7 KB · Views: 99
  • IMG_0976.jpg
    IMG_0976.jpg
    563.3 KB · Views: 133
  • Netvoedingsgedeelte.png
    Netvoedingsgedeelte.png
    19.5 KB · Views: 90
  • Regulator.png
    Regulator.png
    17.1 KB · Views: 128
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.