DIY Audio Projects Forum

I like to use the meter probes that are long and skinny with insulation covering all but the last 1/8 inch of the probe...good for reaching into P2P amps.

_________________
Are we not men?

Basically, test, test, test, and test some more. I do self designed builds. The A Number One thing not to do is slap it together, hit the power switch while praying. Bad idea, and I knew someone once who tried constructing a colour TV kit in just that manner, and he could have bought several with all the money he spent trouble shooting that thing, and it never worked.

Here's how I do my projects:

1) After mounting the hardware, I first wire up the heaters. Check and check again to make sure the right pins are connected. If there are other secondaries, tape these off and get them out of the way. Plug in the tubes, and power up. Make certain all cathodes are glowing.

2) Next, complete the power supply. Again, make certain everything's connected correctly, and pay special attention to the polarities of all electrolytics. So far, I haven't lot a one, but found several with reversed polarities occasionally. If using solid state, also make certain the diodes are connected for the correct polarity. If this is a solid state supply, then a 100W light bulb in series with the PTX primary is a really good idea. It should light up brightly for a second or so as the filter capacitors charge up, then settle to a dim glow. At that time, measure the DC voltage(s). They'll be low, but should be there in proportion to the actual primary voltage. If that looks good, try it at full power. Hollow state PSs with integral diode heater secondaries can't be tested this way, since that'll make the heater voltage too low, and cold cathodes compromise both PRV and Isurge ratings. You could get a nasty flash-over with an abnormally cool cathode. In that case, a clip on AC ammeter is advisable to make sure the primary current is in line with the expected value. If it's excessive, then something isn't right, and you need to find out what that is before powering up.

3) Begin wiring the circuit itself. If using any solid state CCS's as active loads, it's best to complete these first and test to make sure they're working as expected. I use 9V batteries connected in series for this, and variable resistors to make sure the current really is constant, and within design nominal values. I've found this is especially important when using those multipacks of diodes from Rat Shack, as I've seen bad ones in there. Same goes for those "transistors anonymous" packs they sell.

Complete the first pre, always checking the wiring for correctness, and power it up. O'scope to check for undistorted output, voltage gain, and check for parasitics. Make sure its right before proceeding. Test every stage as it's completed to the finals.

4) If it's a design that includes feedback loops, be certain to connect a large load resistor to the output because Mr. Murphy is certain to have connected it for positive feedback more often than not, and you don't want to pump max watts into speeks when it behaves like a 50W Royer oscillator. O'scope and make certain that there are no signals at both the secondary and primary of the OPT that shouldn't be there. Depending on the finals, you sometimes catch snivets at the primary (looks like "bubbles" along a sine wave when testing with a signal generator).

When doing solid state projects, always bring up the voltage with a variac. Transistors just love to oscillate at radio frequencies, and when they do they poof faster than any fuse can blow. You need to catch that before it can build to a level that'll poof the finals.

_________________
Commercial amplifiers may not have been designed to be the best, but they were certainly designed to be the cheapest.

I may have overlooked this advice in the earlier responses but I always use an emergency stop with a (red coloured)mushroom head.
This stop I have mounted on a piece of wood and placed it under my lab table as near as possible to my feet.
This switch interrupts the mains voltage to my test chassis. And in case of disaster, I only have to push the mushroom head with one
foot to switch off the mains voltage.
It is really a life saving set up!

Mpfffff..... Lots-O-Good stuff here. But, not (in my opinion) distilled into a single process. I will *try* to do so, and I also invite any-and-all suggestions or improvements.

First, I will describe my set-up for first test - together with a little background:

I came into the audio hobby by way of vintage radios. I always enjoyed audio, but as a purchaser-of-stuff vs. a builder-repairer-of-stuff. That came as a natural outgrowth of the vintage radio hobby. Most vintage radios in my collection are tube-based, some go back as early as 1919, but as to mains-powered radios, to 1928. Filter caps of that vintage, electrolytic caps of any vintage other than yesterday, and such are always questionable, and often shorted. "Just Plug It In" lets out the magic smoke in short order.

From the wall, I have a 1900 box that includes a regular duplex receptacle and a 15A GFIC device. This provides ground-fault protection if necessary. That, in turn, feeds a Heathkit IP5220 metered ISO(!!)-variac with two ranges 0-1A and 0-3A. For the record, I have only one audio device that exceeds 3A in regular operation - that would be my Citation 16 amp.

So, when a new-to-me device comes to the bench, I will plug it into the Heath and start at 0-volts. As I bring it up, I keep an eagle-eye on the ammeter, if it starts to spike, I can turn everything off well before any damage is done, or any smoke escapes. Further to that, each range is separately fused, adding a small additional level of protection. After a number of years using this device and observing the behavior of various components on power-up, the process becomes predicable such that when it deviates from the proper path, it becomes obvious.

Example: Most well-behaved items with tube rectifiers will show a small current draw increasing in a linear fashion to the voltage *until* the rectifier starts to pass DC. Then, there will be a fast, but orderly ramp-up until full operating voltage is reached and the filter caps are charged. This IS predictable with a narrow range based on the type of device, number of tubes, type of tubes and some other basics. If the current passes that predictable range, or "hunts" - there is trouble. And with SS rectifiers, with each voltage rise, the current draw will jump, then fall back, jump, then fall back to the proper range, all things being good.

So, at the very least, this will allow any new device to be tested at a very low risk, and even save fuses as the device should never reach the point where a fuse will be blown. I cannot imagine doing it any other way with any degree of safety. I cannot imagine testing anything without the ability to measure current within a very few watts. And, for the record, a 1936 6-tube Zenith radio that crossed my bench a few weeks ago came up on the Heath drawing 160 watts at 105V, and actually played with only a little bit of hum.... and within 30 seconds of noise, that needle started to creep up yet more (all predictable after 30 years in the hobby - so going to 105V was only to find that 'creep' point). After recapping, cleaning and some other minor repairs, it pulls about 35 watts quiescent. I wonder where that extra 125 watts was going (not really)?? Point being that the moment a 6-tuber exceeded 50 watts, I knew I had a problem.

Cutting to the chase: Critical to initial power-up, one needs:

a) A means to protect the user against ground faults - GFIC, at least.
b) A means to isolate the user from Line-Hot or Line-Neutral - Isolation transformer.
c) A means to control voltage accurately.
d) A means to measure current accurately and in tiny increments.

However one does it, these are the basic means that MUST BE included for both basic safety and then protection of the test piece. Anything less puts one or both at risk. There are any number of ways to get there, but get there one must if one chooses to indulge in this hobby.

I haven't read the proceeding posts but here is my method for turning on tube amps for the first time; Absolute number one, check and re-check your wiring and compare it against the schematic, use at least two meters - an analogue for the HT and digital for the filament voltage. AND clip lead to the test points, no fingers in the machinery.

When the power comes up and be ready to pull the plug if voltages etc. are not within expected range. Also look away on power up. If a cap goes off you could lose an eye.

But careful wiring and double checking that against schematic can go a long way to a safe maiden voyage.

_________________
Projects: "Lagoon" - tube preamp with cathode follower | "retro-Oatley 6J6" - 6J6 push-pull headphone amp with OPTs | "retro-Hiraga" - Jean Hiraga Le Monster
Website: retro-thermionic

No one had recomended this, but I recomend wearing plastic gloves whenver the amp is turned on.
Some sort of thinn latex glove are good, dosen't stop you from manurvability but gives you protection from high volts.