Can you (or someone) give some guidance around what would be a good variac? Any particular specs or brand? Thanks.
It is rumored that Steve at ApexJr still has several variacs on his shelf, but not listed on the website?
Get a set of Spotting Drills for drilling holes position perfect. When countersinking into Aluminium, use a Hole-type Countersink bit. I have tried all the tricks with the Fluted types but they still chatter.
Great advice. I have to do this for where the miniclips cant enter 🙂
I'd add to solder resistors with more space over the pcb so the miniclips would fit and stay anchored 🙂
1. Build your power supply first. Test it, using a bulb tester and confirm your "no load" voltages.
2. Build one amp channel next. Ground the inputs for your first tests. Do NOT test with an expensive loudspeaker system. Connect it to the power supply and test. Use the bulb tester at first. (Read, don't blow out TWO channels of an amp, just because they both have the same wiring error!)
3. Build the second channel. Wire it the same as the first. Connect and test.
4. Don't tell you're "significant other" how much you are spending on transformers, heatsinks and your F5X projects......!! 😱
2. Build one amp channel next. Ground the inputs for your first tests. Do NOT test with an expensive loudspeaker system. Connect it to the power supply and test. Use the bulb tester at first. (Read, don't blow out TWO channels of an amp, just because they both have the same wiring error!)
3. Build the second channel. Wire it the same as the first. Connect and test.
4. Don't tell you're "significant other" how much you are spending on transformers, heatsinks and your F5X projects......!! 😱
Measure twice, cut once... And... Hundred times of regrets!
Sorry for offtop but sometimes it is most real thing that happens)
Best regards, Anton
Sorry for offtop but sometimes it is most real thing that happens)
Best regards, Anton
On the Flukes (handhelds) that I'm used to using, to measure current you not only have to switch to the 'current' setting on the selector switch, but you must also swap the positive lead from the 'V-ohms-diode check' input to the 'A' input. In this configuration, the meter is basically a dead short lead to lead, and (in my case anyway) fused at 10A.
If, without paying enough attention, you measure current and then later switch to voltage and go to take a measurement without swapping the lead back to the V jack, you wind up putting a short that can carry 10A across whatever you're trying to measure the voltage of. Something then frequently goes 'poof'. It may be the fuse in the meter, or it may be something in the circuit you're measuring that doesn't like short circuits.
I could say I've never done that, but it would be a lie.I'm guessing that your Keithley doesn't require the input to be swapped for current measurements. Is it a bench meter, rather than a handheld?
-Pat
Last edited:
One of the big problems that comes up time and time again is hum because of grounding problems.
The usual cause is mixing power and analogue grounds.
Under no circumstances should these be mixed, but they should join at one star groun point.
I designed a pcb for a audio mixer and the hum was terrible.
I just laid out the pcb with as short a ground as possible.
The power supply was being charged up from the transformer via a bridge but these charging pulses puts 50Hz into the power supply ground.
The 50Hz was modulating the audio ground and causing massive hum.
I relaid out the pcb with the power supply seperate from the audio.
The grounds met at one point only at the connector to the pcb from the power supply.
This time the hum was extremely low and I had difficulty hearing it.
The usual cause is mixing power and analogue grounds.
Under no circumstances should these be mixed, but they should join at one star groun point.
I designed a pcb for a audio mixer and the hum was terrible.
I just laid out the pcb with as short a ground as possible.
The power supply was being charged up from the transformer via a bridge but these charging pulses puts 50Hz into the power supply ground.
The 50Hz was modulating the audio ground and causing massive hum.
I relaid out the pcb with the power supply seperate from the audio.
The grounds met at one point only at the connector to the pcb from the power supply.
This time the hum was extremely low and I had difficulty hearing it.
There are two things I do (as a beginner myself) that seem to help greatly when populating a PCB.
First, I measure every passive component before it goes in. Verify that the 1K resistor is really 1K, and that the 10pf cap is really 10pf. I've found several near misses this way.
Second, once I've soldered a component in (any from resistors, to transistors, or RCA jacks) I measure continuity across the PCB. To do this, set your multimeter to beep for continuity. Place one probe on the lead of your component (before the solder joint) and the other probe on the next location on the PCB. You may have to inspect both sides of the board to find a suitable probe point, but shouldn't be a problem unless you are using a multi-layer board (unlikely for beginners anyway).
This does take quite a bit of extra time. However, it's certainly less time that troubleshooting an unknown error that could have been prevented. Plus, this is my hobby - I don't care how long it takes <g>
First, I measure every passive component before it goes in. Verify that the 1K resistor is really 1K, and that the 10pf cap is really 10pf. I've found several near misses this way.
Second, once I've soldered a component in (any from resistors, to transistors, or RCA jacks) I measure continuity across the PCB. To do this, set your multimeter to beep for continuity. Place one probe on the lead of your component (before the solder joint) and the other probe on the next location on the PCB. You may have to inspect both sides of the board to find a suitable probe point, but shouldn't be a problem unless you are using a multi-layer board (unlikely for beginners anyway).
This does take quite a bit of extra time. However, it's certainly less time that troubleshooting an unknown error that could have been prevented. Plus, this is my hobby - I don't care how long it takes <g>
- Status
- Not open for further replies.