Phantom from a boost converter

[abbey road d enfer]

I've run tests,

That would indicate that the current injected in the base of the transistor is either too small, or too high, probably the latter. Have you put a resistor between base and ground, to discharge it?
Can you monitor the current, with a 0.1 ohm emitter resistor?

 

[KA-Electonics.com]

This is a good reference for boost converters using the MC34063: HV Supply Kit

I've used his kits for Nixie supplies.

 

[saint gillis]

That would indicate that the current injected in the base of the transistor is either too small, or too high, probably the latter. Have you put a resistor between base and ground, to discharge it?
Can you monitor the current, with a 0.1 ohm emitter resistor?

Yes a 3k3 between base and ground.
And I've measured about 5mA injected in the base

 

[saint gillis]

This is a good reference for boost converters using the MC34063: HV Supply Kit

I've used his kits for Nixie supplies.

That's a very interesting resource, thanks a lot !
However, the number of components start to make the project less interesting than using an LM2577-ADJ or an MIC3172 considering pcb space / price ..

 

[abbey road d enfer]

And I've measured about 5mA injected in the base

Peak or average?

 

[saint gillis]

Peak or average?

average

 

[saint gillis]

~250mA peak..

 

[saint gillis]

Would it be possible that the transistor needs some kind of special requirements to be used in a switching application like that ?
What happens here is brief but quite violent isn't it?

 

[Hubbub]

~250mA peak..

View attachment 112089

Given your 200ns/div sweep choice this isn't a fair picture of the base current (or voltage) waveform you have, 2 us or 5 us would do better. At the switching frequency - around 100KHz for an MC34063 - it should be a fairly clean rectangle, with about a 90 % on time if boosting 5V to 45V. The collector current should peak out at about 9-12 times the load current just at when the base drive drops. A safe design has the base current about equal to the expected max load current, you may have the pin 8 resistor too low. When the base is driven the collector (~emitter) current should ramp some, but if much over a 25% change the inductor is too low an inductance value. If the current ramp isn't a mostly straight line and hooks up at the end that means the core is saturating and the inductor needs a higher current rating.

 

[thor.zmt]

New thread inside the thread :
Here's a boost converter from MC34063 datasheet :
View attachment 112028

Here's the circuit recommended to provide more current with an external transistor :
View attachment 112029

How would you do if you wanted to use an external transistor to avoid more than 40V going into pin1 (as the MC34063 doesn' accept more than 40V) AND getting 48V at Vout ?

Don't use a simple Boost converter, add a charge pump to double the output voltage.

This way you generate as much as ~ 80V DC out from any sensible input.

Pump it up with charge pumps – part 3

With MC33063A you could for example get +24V and +48V from a 5V - 12V DC input.



Thor

 

[thor.zmt]

I would try something like that. With a 12-20V zener, making sure the reverse voltage is high enough. No gurantee here.
View attachment 112030

Not recommended. You could make a suitable regulated PSU for Pin 1. As shown elsewhere, a charge pump doubler (or tripler) is probably the better choice.

One MC33063A could in theory produce +/-18V & +54V (and say -24V Bias for J-Fet switches in the signal path) from one IC with (say) 7.5W input power from 5V. These can run from commodity USB phone / laptop chargers. More modern USB-C chargers can use USB Power Delivery (USB-PD) with voltages of 5/9/12/15/20V at 3A, so a lot more power could be produced.

Here the principle using a different switcher IC.



When doing this I personally used IC's more than 10 times faster and more modern, having moved the past MC33063A rapidly. But if space is not a huge premium, why not use MC33063A?

Switching at 100kHz still put's us way past audio and other than capacitors and inductors are bigger and through hole (which is good for DIY) and cost a bit more there is no real drawback.

Use Panasonic "Os-Con" capacitors as filter and reservoir capacitor - they have super low ESR and are worth the extra. Normal Solid Polymer capacitors are recommended for other positions.

Thor