"Bringing up" an amplifier

ask your general tube related questions here

"Bringing up" an amplifier

Postby erichayes » Thu Sep 02, 2004 11:06 pm

Hi All,

I thought I'd relate the techniques I've honed over the last 40+ years for getting an amplifier or radio that's been unused for several years operational... with a little luck. There is no guarantee that the procedures will yield 100% success, but they do greatly improve the odds.

Rather than just Heathkit you blindly through the procedures, I'll attempt to explain why you're doing what you're doing; this, hopefully, will stimulate your imagination and intuition and enable you to develop your own methods of restoration.

The required test equipment for proceeding with this adventure are a variac, or variation thereof--see posts on other threads here--some clip leads, a 100 ohm 10 watt resistor and at least one sensitive voltmeter capable of reading 600VDC, either a DVM or a VTVM. Two will make things a lot easier. An analog VOM is not recommended as its internal resistance will load down the circuit, and false readings will result.

Also highly recommended are solid state rectifier substitutes for the vacuum tubes probably present. While prebuilt substitutes are available for 5U4/5Y3/5AR4 type tubes, they're relatively expensive, and don't cover a bunch of other rectifiers. I recommend buying a handful of 1N4007s, a couple of octal male bases, and, if you're into old radios or jukeboxes, a four pin base. Antique Electronics Supply has these parts available (P-Q1N4007, P-SP8-476, P-SP4-477 respectively) for very reasonable prices. This will allow you to substitute solid state diodes for virtually all tube rectifiers. The reason for using silicon instead of thermionics is simple: rectification starts as soon as you start applying AC voltage to the set. With a tube rectifier, the cathode must reach a certain thermal threshold before it starts emitting electrons. This usually starts at around a 35 volt AC input for a directly heated cathode (5U4/5Y3 etc) or 60 VAC for an indirectly heated cathode (5AR4/6AX5/6X5 et al), and the DC voltage rises rapidly as soon as this threshold is reached. I should have mentioned this earlier, but the two key words in bringing up an amp or radio are "control" and "patience". By allowing the tube rectifier to warm up arbitrarily, you've lost control of the situation; solid state diodes allow you to monitor the whole run-up and make corrections before it's too late.

To make the octal substitutes, take a couple of 1N4007s and insert one of the anodes (unbanded end) into pin 4 of the base, and the anode of the other into pin 6. Wrap the cathode lead of one of the diodes around the cathode lead of the other, solder, and clip. Insert the remaining cathode lead into pin 8 of the base. Don't forget you're looking down at the tube base, rather than up at the tube socket, so the pins are numbered counterclockwise, i.e., pin 1 is to the right of the key. This is your sub for the 5AR4, 5U4, 5Y3, 5V3 family of rectifiers. Do the same thing with the second base, but use pins 3 and 5 for the anodes. This takes care of the 5X4, 5Y4, 6AX5, 6X5 and (if you're working on an old car radio) 0Z4.

Miniature tube rectifiers are a bit more tricky, as there are no 7 or 9 pin bases available. What I do is build up the lead diameter of the diodes with layers of solder until they make a snug fit in the tube socket. Not too snug, or you'll wind up trying to recompress the socket pins with a dental explorer or jeweler's screwdriver. Again, tie the cathodes together and, in the case of 6X4s or 12X4s, insert the anodes into pins 1 and 6, and the combined cathode lead into pin 7. In the case of EZ80/6V4 or EZ81/6CA4, the anodes go to pins 1 and 7, and the cathodes to pin 3.

For an 80, 83 or 5Z3 sub, insert the anodes into pins 2 and 3 (the skinny pins of the 4 pin base) and the cathodes into pin 4 or pin 1.

Of course, you can eliminate all of the above tedium by simply tack-soldering the diodes to the tube socket, but if you're going to make a career of restoring old stuff it's nice to have a few conveniences.

Okay, let's assume you've assembled the above-described apparati and are ready to start the run-up. First thing to do is remove all tubes from the set undergoing treatment. The reasons are:

1. You don't want the set trying to operate as you bring it up, as there might be other hidden problems that you'll have to address.

2. You don't want to be looking at any current draw other than that of the filter network.

3. Since we're dealing with potential, rather than power consumption here, the rated voltages of the various caps can be reached with less line voltage applied.

Second, check the voltage ratings of the filter caps--not the cathode bypass caps, which are usually rated at 25~50WVDC--and note the lowest voltage. It'll probably be 250 to 350 WVDC.

That's it for chapter one.

Chapter two: Conventional full wave rectifier power supplies.

First thing to do is visually inspect all the filter capacitors for leakage. if there is a white or blue-green crystalline "growth" around any terminal or lead of the cap, replace it. Period.

Before applying power, remove the rectifier tube, or unsolder the solid state diodes from the input of the filter network. This is usually a cap around 40 µF, although choke-input filters are not unheard of in consumer electronics. There may or may not be a low value resistor between the rectifier and the first filter leg. We'll be going back to this in a moment.

Take your meter, set for the highest resistance range, and measure the resistance from the first filter leg (be it choke or cap) to ground. You should see a very high to infinite resistance, as most manufacturers don't build bleeder resistors into their circuitry for economic reasons. If you get a reading of less than 100 KΩ, check the schematic for a bleeder, or physically inspect the circuit to see if there's a resistor going from one of the filters to ground. If there is, disconnect it. If there isn't, go to each filter cap and check its resistance to ground. The cap with the lowest resistance will be defective, but it might not be the only one. Disconnect it from the circuit, and repeat the procedure. NOTE: the cap most likely to fail, either by opening up or becoming leaky, is the primary cap. It's having to deal with raw rectified DC, which has a considerable AC component present. This, to put it mildly, is a rotten environment for a capacitor.

If the network passes the resistance test, it's time to reconnect the rectifier and apply power. Referring to the above mention of a resistor between the cathode and the first cap, if there is no resistor, cut the wire between the cathode and cap, and solder or clip-lead the 100 Ω resistor across the cut.

Using whatever variable voltage supply you have (and I still strongly recommend a real variac), and with your meter set for AC volts and connected across the line cord of the subject set, adjust the input voltage to 30 VAC.

Remove the meter and reset it for DC volts at the highest range (or autorange if it has it). Connect it between the cathode of the rectifier and ground, and note the reading.

This is where it starts getting a little tricky, but don't quit now. We've already discovered (and, I assume, replaced) any bad caps using the resistive test, but now we're applying voltage to the caps. Whole different critter.

Reconnect the meter to the other side of the resistor between the cathode and cap--the cap side. There should be virtually no difference between this reading and the one you got off the cathode. If there is, DO NOT INCREASE THE AC INPUT VOLTAGE. This is where patience comes into play. Watch the voltage, and see if it starts dropping. If it does, that means one or more of the caps is starting to re-form.

Because all the tubes have been removed, the only way current can flow is through the filter network, itself. If there's a voltage drop across that first resistor, something downstream is conducting, i.e., a leaky cap. By running it at substantially reduced voltage, the electrolyte might be able to reconstitute itself without arcing through to the adjacent foil. Emphasis on "might". If you're showing current flow at this low voltage, you should probably throw in the towel and recap.

Assuming the current flow drops to near-zero, advance the input voltage slowly, while monitoring the DC at the rectifier and the voltage drop across the resistor. Yeah, that means you're going to be jockeying between two test points; I told you two meters would diminish grief.

If you're able to increase the input voltage and see favorable results in the voltage drop on the resistor, it's time to start looking at the absolute DC voltage. If it's approaching the maximum voltage value of the last cap in the string, you're going to have to disconnect it (or them) before you increase the input voltage. Make a note of the input voltage, shut the jig down, and do the necessary surgery. Repower the set starting from zero input volts, but increase the voltage more aggressively while watching the voltage drop across the resistor. It should rise and fairly rapidly fall as the AC voltage is increased. If it doesn't, let the set soak at whatever voltage you see the fall rate become sluggish.

Chapter three tomorrow.

Chapter three--two days late--: Voltage doublers.

Up until now, we've been discussing conventional full wave rectification using a center-tapped AC source. There are two other rectifying techniques, both almost exclusively involving solid state diodes in consumer equipment: the full wave bridge and the voltage doubler.

The full wave bridge, or FWB, can be treated essentially the same as the center tapped full wave design; the only difference is the DC output (with a cap input filter network) is 1.414 times the AC input instead of .707 times the total AC input for a conventional FWCT. There are some minor considerations involved between the two, but we can nitpick later.

The voltage doubler, on the other hand, is substantially different, in that it is relying on a relatively large capacitance to act as a "phantom" secondary winding of the power transformer. Simply put, two diodes are connected to one end of the high voltage winding with opposite polarity, i.e. cathode of one diode and anode of the other. The other ends of the diodes are connected to relatively high capacitance, medium-high voltage caps--usually around 200-300 µF @ 250-300 WVDC. These caps are wired in series, and the other end of the transformer winding is connected to the junction of the two caps. The positive lead of the "upper" cap is B+ and the negative lead of the "lower" cap goes to ground.

During the positive excursion of the AC wave, the upper cap charges; during the negative excursion, the lower cap charges. Since the two caps are in series, the total voltage developed across them, ideally, is twice what we'd see in a conventional FW circuit. Please note that there's no such thing as a choke input voltage doubler in the real world.

The only consideration that has to be made with a voltage doubler is that the input caps are of equal value. This is easily determined by going through the bringing-up procedure previously described, and checking the voltage across the upper cap and comparing it to that of the lower cap. They should be within a very few volts of each other. If they're off by more than 10 volts, change them both. The rest of the startup routine is as described earlier. BTW, an easy way to determine if you're dealing with a VD supply without even popping the bottom plate is to see if one of the above-chassis can caps is insulated with cardboard. This will be the upper cap, and will have a potential of ½ the total B+ on the can--hence the insulation. BE SURE TO INSULATE THIS CAP'S EXTERIOR IF YOU HAVE TO REPLACE IT.

Questions?
Last edited by erichayes on Thu Aug 23, 2007 1:27 am, edited 1 time in total.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

RE: "Bringing up an amplifier"

Postby Norris » Fri Sep 10, 2004 8:16 pm

Hi Eric,

I want to thank you for going to a great deal of effort in writting this information about bringing up amplifiers.
This is well thought out and detailed to the point, that even a beginer like myself should be able to get it right.
I am glad that there are people such as yourself and others here that are willing to share their knowledge and time to help ones who do not always understand.

Thanks again for all of your help.

Norris Wilson
Norris
 

Postby erichayes » Thu Aug 23, 2007 1:01 am

Hi All,

Thought it might be appropriate to dredge this thread up, given Mike's comment on the can cap going boom.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby parabellum » Thu Aug 23, 2007 9:28 pm

sweet ... I never saw this thread
User avatar
parabellum
 
Posts: 195
Joined: Wed Apr 25, 2007 1:37 pm
Location: Bay Area, CA

Postby Vince » Fri Aug 24, 2007 1:49 pm

Thanks Eric! Very timely.
I am about to start on the Lowery Organ amp and I would have screwed it up
big time without this for guidance.
My first piece of test gear is a Fluke 8012A. For the second I dug out a PACO
Model V-70 VTVM. I opened it up and it has a selenium rectifier and the 12AU7A is not lighting up. Luckily, I had picked up a manual for it a couple months ago. But having to fix it first is a royal PITA.
Thanks again for the re-o=post, Vince
Vince
 
Posts: 118
Joined: Thu May 24, 2007 11:34 am
Location: SF Bay Area

Re: "Bringing up" an amplifier

Postby Ty_Bower » Mon Oct 01, 2007 8:16 pm

erichayes wrote:First thing to do is visually inspect all the filter capacitors for leakage. if there is a white or blue-green crystalline "growth" around any terminal or lead of the cap, replace it. Period.

...Questions?


Is this one dead, Jim?
Image

These are the CE caps, 80/40/30/20. I put the analog VOM on them. The sloppy looking one (on the left) ohms out to 270k/100k/400k/450k per section respectively. His brother (on the right) is about 1000k/180k/>10meg/>10meg.
User avatar
Ty_Bower
KT88
 
Posts: 1492
Joined: Wed Mar 21, 2007 2:50 pm
Location: Newark, DE

Postby erichayes » Mon Oct 01, 2007 11:35 pm

I'd be real leery of the one on the left, Ty; those numbers are nervously low. The other one might be okay, but I think a call to AES would be in order regarding both of them. It's been a few years, but I got defective CE cans and they replaced them without any hassle. I'm assuming their policy hasn't changed.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby erichayes » Sun May 04, 2008 1:04 am

Here it is . . . again. Sorry to be repetitious to the "older" folks.

Shannon, whatcha think about creating a "library" topic, wherein the posts made by the old farts to guide the ignorant intelligencia could be corralled into one category.

I know "Tube 101" was implemented originally for that purpose, but holy crap, Shannon, this forum has grown exponentially since its inception (take a well deserved bow), and it's a little silly to keep reposting stuff that can be archived . . . oh, sh!t, I think I just spoke one word too many.

My carpal tunnel is going to come back tomorrow and bring me to my knees.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby erichayes » Tue Sep 30, 2008 2:00 am

Once again, I'll drag this tome out of the archives.
Eric in the Jefferson State
erichayes
KT88
 
Posts: 987
Joined: Fri Jan 23, 2004 9:01 pm
Location: McKinleyville CA

Postby Kojak » Thu Oct 02, 2008 8:08 pm

Thanks Eric-

Great 'article', saved it to my hard drive.

best, Ed R
User avatar
Kojak
 
Posts: 45
Joined: Fri Feb 22, 2008 10:59 pm
Location: Nashville, TN

Postby nyazzip » Sun Oct 12, 2008 5:33 pm

being the hack that i am, immediately after i soldered in my new power supply capacitors on my '74 traynor, i forgot about the "bringing up with variac" wisdom i had actually read before(but forgotten). anyway, i intoduced full line voltage immediately, then after a minute or so switched from standby to operate....no problems. still, i wish i had been smart enough to bring 'em up gradually, as i do actually own a variac. oh well. funny thing is i have had the top off my amp all summer long, so if the caps wanted to go like a grenade, there'd be nothing stopping the shrapnel
User avatar
nyazzip
KT88
 
Posts: 1070
Joined: Wed Jan 09, 2008 1:24 am


Return to tube 101

Who is online

Users browsing this forum: No registered users and 1 guest