This is a mains frequency switcher. It uses the leakage inductance of the transformer to limit the capacitor charging current. That’s why the sine wave “flattens out” when the capacitor is connected via the SCRs. When the capacitor is connected, the voltage across the secondary is the capacitor voltage. We are measuring the voltage across the secondary + leakage inductance so we don’t get to see the voltage going to that leakage inductor. We can see it by looking at the primary waveform, or by winding a couple voltage sense turns on the transformer to avoid dealing with mains. A current probe would show very nicely that the time derivative of the capacitor charging current follows the voltage across the leakage inductance. The transformer is probably designed to have sufficient leakage inductance, otherwise the concept won’t work. This is also why in this supply if the transformer fails, it should be rewound identically and reuse the magnetic circuit. That way the leakage inductance will be maintained. The transformer is buzzing because the secondary current is very rich in harmonics - much more than it would be if the SCRs were permanently turned on (ie if we had just a plain old rectifier bridge). The buzzing could be eliminated by allowing a bit more losses on the rectifier, thus replacing SCRs with mosfet pairs, driven by a dI/dt and dV/dt regulator.
Your power supply is possibly an export model, designed for the New Zealand market. We get a lot of earth quakes here, so it's good to have instruments that work well after they've fallen over. Maybe it will work even better if you do a 600mm drop test to simulate what happens after a magnitude 6.5+ shake, when things fall off the bench onto the floor.
For giggles, try wiping the meter face with a laundry anti-static cloth and make sure that's not your problem. I've seen this more than once over the years.
I've got a few of these old HP guys. My meter switches maybe also need a regular clean. Some of the "precision" guys use a zener in a can with heater. Woo! I never got WHAT the heck the SCR's were in the bridge for! NOW I know! Thanks! 73's, W3IHM
Yeah, sticky meter movements will trip you up every time. I'm glad the power supply is working otherwise. I noticed that power transformer secondary is jerked down to the voltage of the bulk filter cap when the SCR fires. This must make for terrible power factor and EMI. I noticed on some of the earlier Harrison pre-regulator circuits, they used a pi input filter on the line for this reason.
Yes, the battery operated scope shows the power regulation bite taken out of the sine wave nicely. I tape notes to my projects to remind me of their status, when I return to them. A digital meter might be a nice upgrade to the unit... It would probably be easier accomplished and less expensive than a new analog meter, as well. A MCU could add a lot of functionality, also.
The pre-regulation is interesting, it switches on after the the voltage maximum in the case shown so that the capacitor stays charged at a voltage just sufficient for the main regulation. As a consequence the angle of current flow also gets smaller. So for low voltage and high current at the output the peak current in the transformer gets really high, maybe 30A or more for 10A at the output but I am sure the transformer is strong enough for the operation.
Die ordnungsgemässe Funktion eines Drehspulinstrumentes ist *lageabhängig* . Stell das HP-Supply auf seine Füsse und die Anzeige wird wieder ordnungsgemäss funktionieren. LG
Another very interesting video, thanks. I have this PSU and have used it for years, probably way above my skill level. I don't short the negative and ground as you have done. Should I have been doing that?
The positive and negative leads were shorted so the current limit could be set. To avoid taking chunks out of the meter leads, I turn the output voltage to zero, short the leads, then turn up the voltage slightly, and then set the current limit.
Yes, if you want a positive supply referenced to ground (typically the case). When you want a negative supply, short the positive terminal to ground. If you don't do this, you'll have a floating supply, which unless you need it, causes trouble when probing the circuit.
As an old technician I know this is heresy but why not use one of those digital meters you have on the smaller supply. You may need a centering plate for it to fit nicely but Brunhilde would be OK with a new digital front panel.
This is a mains frequency switcher. It uses the leakage inductance of the transformer to limit the capacitor charging current. That’s why the sine wave “flattens out” when the capacitor is connected via the SCRs. When the capacitor is connected, the voltage across the secondary is the capacitor voltage. We are measuring the voltage across the secondary + leakage inductance so we don’t get to see the voltage going to that leakage inductor. We can see it by looking at the primary waveform, or by winding a couple voltage sense turns on the transformer to avoid dealing with mains. A current probe would show very nicely that the time derivative of the capacitor charging current follows the voltage across the leakage inductance.
The transformer is probably designed to have sufficient leakage inductance, otherwise the concept won’t work. This is also why in this supply if the transformer fails, it should be rewound identically and reuse the magnetic circuit. That way the leakage inductance will be maintained.
The transformer is buzzing because the secondary current is very rich in harmonics - much more than it would be if the SCRs were permanently turned on (ie if we had just a plain old rectifier bridge).
The buzzing could be eliminated by allowing a bit more losses on the rectifier, thus replacing SCRs with mosfet pairs, driven by a dI/dt and dV/dt regulator.
Your power supply is possibly an export model, designed for the New Zealand market. We get a lot of earth quakes here, so it's good to have instruments that work well after they've fallen over. Maybe it will work even better if you do a 600mm drop test to simulate what happens after a magnitude 6.5+ shake, when things fall off the bench onto the floor.
For giggles, try wiping the meter face with a laundry anti-static cloth and make sure that's not your problem. I've seen this more than once over the years.
WA4 has it right. Static charge on the plastic window. Try it. Easy answer and most always right. Mike in Silver Spring.
I've got a few of these old HP guys. My meter switches maybe also need a regular clean. Some of the "precision" guys use a zener in a can with heater. Woo! I never got WHAT the heck the SCR's were in the bridge for! NOW I know! Thanks! 73's, W3IHM
Yeah, sticky meter movements will trip you up every time. I'm glad the power supply is working otherwise.
I noticed that power transformer secondary is jerked down to the voltage of the bulk filter cap when the SCR fires. This must make for terrible power factor and EMI. I noticed on some of the earlier Harrison pre-regulator circuits, they used a pi input filter on the line for this reason.
Thanks for showing the SCR effects. I was going to ask you to show that. 🙂
Maybe ditch the analog meter and put a digital display in there 😱. A power supply with wrong numbers showing is BAD!
I'd rather fix the analog meter. They have quite a few advantages over those cheap digital ones.
Yes, the battery operated scope shows the power regulation bite taken out of the sine wave nicely. I tape notes to my projects to remind me of their status, when I return to them. A digital meter might be a nice upgrade to the unit... It would probably be easier accomplished and less expensive than a new analog meter, as well. A MCU could add a lot of functionality, also.
Triac phase dimmer circuit for incandescent light dimmers. Have used on old soldering irons to control temp. decades ago.
The pre-regulation is interesting, it switches on after the the voltage maximum in the case shown so that the capacitor stays charged at a voltage just sufficient for the main regulation. As a consequence the angle of current flow also gets smaller. So for low voltage and high current at the output the peak current in the transformer gets really high, maybe 30A or more for 10A at the output but I am sure the transformer is strong enough for the operation.
Die ordnungsgemässe Funktion eines Drehspulinstrumentes ist *lageabhängig* . Stell das HP-Supply auf seine Füsse und die Anzeige wird wieder ordnungsgemäss funktionieren. LG
been there with meters. it is a pain. sometimes bits of iron filing get in the armature -
and then you are screwed.
I've actually been successful removing annoying iron filings sometimes. Some compressed air, magnifier, tweezers, scotch tape.....
Another very interesting video, thanks.
I have this PSU and have used it for years, probably way above my skill level. I don't short the negative and ground as you have done. Should I have been doing that?
The positive and negative leads were shorted so the current limit could be set. To avoid taking chunks out of the meter leads, I turn the output voltage to zero, short the leads, then turn up the voltage slightly, and then set the current limit.
Yes, if you want a positive supply referenced to ground (typically the case). When you want a negative supply, short the positive terminal to ground. If you don't do this, you'll have a floating supply, which unless you need it, causes trouble when probing the circuit.
@@kpnconsulting8739 Agreed!
As an old technician I know this is heresy but why not use one of those digital meters you have on the smaller supply. You may need a centering plate for it to fit nicely but Brunhilde would be OK with a new digital front panel.