Once I have the 4017 working, I can use its output to create a short or long pulse with the
hef4528.

One channel of the hef4528 used to get the B mosfet to switch on at the L2 impulse maximum, which might need a short delay. not sure if it is really needed. definitely a fine tuning attribute.

this way turn on will not have heavy currents as the voltage impulse of L1 will have discharged the dc offset.

the other side of the  hef4528 is used to set the pulse duration for which the B mosfet will stay conducting.
The c2 level is kept at the C4 negative level.
which depends on how long it takes to discharge L1 and C2, and if needed part of the voltage of C6.

this needs to be researched.

the hef4528 will work on 5V again and its output will  drive the B mosfet gate drive IC's

I have one 4017 working, so I can now delay maximum 10 cycles. will now continue with the hef4528.

then if needed if the 4081 arrives I can delay up to 17 cycles (to. charge up the dc offset to higher levels before discharging)

hef4528 pcb prototype is ready for testing

I used 2x 470pF caps
for the first delay to allign with the impulse, Inused 4.7k ohm trimpot
and 100k for the pulse width

caps are in sockets for easy replacement
as is the IC.

Gave it a bnc out for the B mosfets
maybe I'll use a coax cable to connect to the 4017

hmm, L2 C2 are resonant, with low voltage but high resonant current.
I thought this might be a problem.
since the B mosfet discharges this resonant current to ground.
If I prolong the B mosfet on duration, the resonant current will be deminished.
That would also be felt by the L3 secondary.
Cant wait to start testing this.
so much to learn again

as title says.
so now I can make longer delays with 2 or more 4017 counters, to produce higher dc offset before discharging it

During testing the prototype I had a short which I couldn't find.

It might be I shorted an in or output of the hef 4528 to ground instead of to V+

as I used the schematic which gave an inverted output, and I needed a non inverting output.

short should only be present when clock signal is present

hef4528 A input high to low needs B input to be low

A HIGH-to-LOW transition on nA when nB is LOW

 or a LOW-to-HIGH transition on nB when nA is HIGH

 produces a positive pulse (LOW-HIGH-LOW) on nQ

 and a negative pulse (HIGH-LOW-HIGH) on n(Q) if the nCD is HIGH.

Ok, this copy paste doesnt work as I need a line above (Q)

So tying Q or (q) to ground is a bad idea.
instead I should only tie B input to ground.
(Q) is left open
Q and pin 8 provide the output (low high low)

(cd) is always high, else there is no out

fixed the errors in the hef4528
but didn't solve the problem

checked the 4017 which worked fine before, but it definitely is the problem.

without the ic inserted all is fine.

I checked the connections several times and they look fine.

This will be the last time that I build directly to pcb. Breadboard next time... much easier and time saving.

What am I doing wrong? please help!

I build up 2x 4017 (counter) with a 4081 (AND gate)
to be able to count to 17

I get constant output led on pin 3 of the first 4017 while its other outputs stay off

the second 4017 is on at every output.

no blinking leds.

input is a 5v ttl square wave of 1hz

I don't understand why pin 13(enable) and pin 11(last output) have to go to pin 12 of the 4081 input?

13 enables the counter, and should be grounded for the first 4017 to work. I assume it is grounded by pin 11

when pin 11 goes high, then the fiest 4017 stops counting.
at the same time the AND gate 4081 now has a high on both inputs. so the out goes high.
but... there my logic stops.

does the pin 11+13 stay high, until it is reset by pin 15? and if so
How does the clock enter the second 4017?

I used this video as reference

https://youtu.be/JwF_9g8IHUQ

going to put 10k to the AND gate inputs to ground.

I wondered about the power supply being pulsed dc, so also will place a cap there.
maybe even a pi filter if needed for the higher frequencies

ok I removed all, and just build a single 4017,and that works fine.

so chip works.

I swapped in the other 4017
and it doesn't work! so that was the culprit
to the bin yea waste! :emperor:

build up the 2x 4017 and the 4018
and IT WORKS!

pfew...

Wonder why that 4017 decided to die on me.

now continue
 hook it up to the hef4528 and  scope
see if It also works on higher speeds

I noticed my PCB had no 10k from clock im to. ground, that might also have played a role

it seems my buck boost converter is causing the problems.
it needs to convert 12V to 5V and it seems to do just that.

but when I connect the 4017 circuit it girs
 bezerk resulting in failure.

at this time I don't care anymore so I will instead use a lm317 for 5V regulation

another 4017 died.

By itself the 4017 works fine.
but when hooked up to the hef4528
there is a short.

output of 4017 is pulsing +5v and ground

hef4528 is inverted input A on pin 5, so that probably means the ground when led is off,
is directly connected to the hef4528 positive?

if so, then I should probably be able to decouple with a dc blocking capacitor, that passes the high frequency pulse.

input pin 5 also needs a 10k to positive.
so, this becomes a RC filter

pin 11 is the second inverted output, which again needs a 10k to positive

do I again need a decoupling cap for the pin 6 q out to pin 11 a inverted input?
tired... going for a walk.
tomorrow my head will be clear again

nice thanks for posting I like the 4017

https://www.digikey.nl/en/resources/conversion-calculators/conversion-calculator-low-pass-and-high-pass-filter

RC calulator
for 10K ohm and 1nF gives 15kc/s crossover
10nf gives 1.5kc/s

since the first delay only needs to align up with the impulse, I will try to use 100pF. Dont know if that will work, but the RC filter with 10k give 159kc/s which is well below what i need.

This is the circuit. for delaying the Bmosfet
Will test the capacitor values.
I wonder If the first output hef4528 is fast enough, to align the turn on of the B mosfet with the impulse
rp1=5k rp2=100k

the higher the output number, the longer the delay, each delay is one pulse

fingers crossed ready for testing.
again...

I put the two pull up 10k resistors to V+ at the two A inputs of the hef4528 pin 5 and 11

and used a 10nF for pin 5
and a 100pF for pin 11
(dc decoupling capacitor)

will first test with a single 4017.
if it works.... then I can continue builiding up the second 4107 with the 4018.

after which I should be able to soon start testing the B mosfets.

pfew lot of work went into this.
In my head it was simple, but I definitely had a learning curve here.

ok, The idea was good. no more short cuts.
But, now the hef4528 doesnt produce output, due to the small capacity of the first delay,
but more imporantly, the Q output doesnt produce a proper impulse.
so the next stage doesnt produce a pulse.

I already solved this challenge before I measured it :cool:

Instead of using the inverted A inputs, I will use the B outputs.
this has many benefits. I can use the same trimpots and caps for both channels
and I dont need the 2 decoupling caps anymore.

this means I will trigger from the positive going insteasd of negative going transition,
so the fisret channel needs to produce a longer delay (>50% pulse) to allighn the bmosfets with the impulse.
which is fine!
then the second channel does the same trick.

small redesign.

I also worked on the pcb design in EasyEDA, which also will be adjusted.
just small changes so that gives hope...

rewired the hef4528
ready for another test.
will this be the last one?
next time will definitely use the breadboard.

ok hef4528 now works. but I still made a mistake.
The second pulse should be triggered at the end  of the first pulse, but now it is triggered at the start...

the output needs to be at Q  non inverted, for the gate driver IC to work,

so... I need a decoupling capacitor after all.

input is good right now. except, the first delay output should be the inverter Q,  into the B input of the pulse generator.
inverted Q, is active low, so that would mean, so in is high for most of the time.
is that a problem for the B input?

B input is grounded through 10K. if it goes high it triggers.
so the inverted Q out is high, goes low, which is OK I guess.
The the Q inv, goes high again. this is the transistion that should trigger the input.
is it that easy to just connected the inverted output to the B input?

where is my mind!
I should have got this working ages ago. and I still dont get it.
its just so confusing what these logic in and outputs can handle.
I mean, I did have a short before, but I guess that was from the 4017 being grounded. but that would only be possible if the inverted A input was always high.

So lets assume the B input is always grounded... but it isn't or is it.
back to the data sheet

A HIGH-to-LOW transition on A(inv) when B is LOW produces a positive pulse (LOW-HIGH-LOW) on Q

https://assets.nexperia.com/documents/data-sheet/HEF4528B.pdf

ok.. so.. last time.
pint 6 (Q1 out) to pin 11 A(inv in)
pin 12 to ground
pin 11, 10K to pin 13(V+)

ok, I have my delayed pulses working.
The only thing is, my first output is not a proper square wave.
in blue the input clock from the A MOSFET signal, which switches the L1 creating the impulses

yellow is the output, which will switch the Bmosfets, shorting C2 of L2 to ground.

green, is the first delay stage of the hef4528, which makes it able to turn the B mosfets on at the impulse maximum.
And as can be seen, it is not a proper square wave. maybe I need another 10k to ground on the Q output?

I also measured the extremes, which seems more than enough for testing.

happy it works, just need to get that output pulse proper.