Test and Measurement Forum

The DrDAQ HV PSU - Weekend special

Post all DrDAQ questions here

The DrDAQ HV PSU - Weekend special

Postby Glovisol » Sat Jan 17, 2015 8:26 pm

PSU Schematic.jpg
Fig. 1 - Wide Range PSU schematic diagram
PSU 1 - DC 95.jpg
Fig. 5 - System @ 95% D.C. -Output exceeds 100V
PSU 3 - DC 10.jpg
Fig. 3 - System @ 10% Duty Cycle - Note parameters display
PSU 0 - Linear UpRamp.jpg
Fig. 2 - Linear 3V UpRamp, later adjusted to 2V
This is an example of the notable versatility of the DrDAQ in the lab: the unit, complete with its buffer and a few extra components out of the junk box, let me build the prototype of a 400 hz Switch Mode supply in a Saturday's afternoon. The real bonus is that the PSU, by means of Picoscope, is fully controlled and monitored on the PC screen.

The operating principle is quite simple. The DrDQ + Picoscope generates a linear, 2V pp UpRamp at 400 Hz. The standard UpRamp is symmetrical about 0 V and needs to be offset by 1 V: we now have an UpRamp from 0 V to 2 V. We feed this signal to the non-inverting (+) input of Buffer's Amplifier B. On the inverting input (-) we feed a D.C. control signal variable beween 0 and + 2 V.

Suppose the control signal is 1 V. As the UpRamp starts, the Op Amp (-) input sits at +1V and the (+) input sits at 0V. Op Amp output is 0V. Then the UpRamp level gradually rises: when it overtakes 1V, the Op Amp switches to + VCC and stays there until the UpRamp reaches 2 V and then falls abruptly to zero bringing also the Op Amp's output to zero. We have just described the variable PWM generator at 50% Duty Cycle. If the control signal is 0.2 V we shall have an output pulse at 10% Duty Cycle and so on. If this pulse drives a Bipolar, or a MOSFET or an IGBT switch working in a BOOST type flyback converter, you have a wide range, low to high voltage adjustable PSU.

Fig. 1 shows the simple diagram. Control voltage is adjusted with R 103. I had to make do with what I had on hand, hence the 220V to 15+15V transformer for the flyback coil and the 2N3055 Darlington connected switch. Nevertheless the "beast" puts out a respectable 90V @ 0.5 A while drawing 6 A from the 13V supply. Therefore we have a wide range unit that can be continously adjusted from 15 V up to 100 V! The screens show the system operating with a 3V UpRamp, but later I set the UpRamp at 2V, as described above. Reason for this is not to exceed the 2.5 V dynamic range of the EXT. inputs. By changing the UpRamp frequency one can easily check and compare performance. I quickly and easily tested the system from 200 up to 2 KHz. The screens also show the full monitoring capability of the DrDAQ/Picoscope combination. The SCOPE scale is colored GREEN and shows the switching waveform. The PURPLE scale on the left shows EXT. 3 Channel monitoring the UpRamp (here still at 3V). The UpRamp looks bad because of insufficient channel bandwith. The BROWN scale at right shows the level of the CONTROL D.C. voltage on EXT. 1.

The system shown works open loop. In a next post I shall describe results obtained by closing the loop, thereby obtaining a self regulated PSU.
PSU 2 - DC 50.jpg
Fig. 4 - System @ 50% D.C. Deformed ramp because of low bandwith of Ext. 3 input
Advanced User
Advanced User
Posts: 248
Joined: Sat May 11, 2013 10:03 am
Location: San Marzano Oliveto, Asti, Italy


Postby Glovisol » Wed Jan 21, 2015 1:58 pm

PSU Design principles.jpg
Fig. 6 - Schematic of the closed loop PSU - connections to the DrDAQ not shown for simplicity
For loop control an additional Op Amp is required, to act as error amplifier. Since I did not have another spare amp in the Buffer, I decided to use a twin amp LM358, where one would be the comparator and the other the error amp. In this way the spare Buffer Amp became free again. To verify the ripple suppression of system in closed loop conditions, I built a raw supply as well (it was unfair to use a regulated and ripple free supply for the source, as I did in Open Loop Mode). The complete unit is shown in Fig. 6 and in the photos. Once loop calculations are done as per attached EXCEL file, setting it up is quite straightforward.

When we were in manual control, the input/ output rule was measured as follows:
V0 = 20V......................Vc = 1.60V
V0 = 50V......................Vc = 0.56V
V0 = 100V......................Vc = 0,18V

As explained in Fig. 6, Variable VR (VReference) was chosen. With a reduction ratio of 50 (e.g. 100 V become 2V at the error amp input) the necessary VR voltages are readily calculated. In this way output voltage V0 is set by setting VR with 10 Turns pot R6. Loop gain is determined by the ratio R4/R3. To set it up, initially make R4 = 3.3 KOhm for unity gain. In this condition the system will behave as open loop because output voltage will have negligible effect on Duty Cycle. Connect suitable load to output (I used a filament lamp) fire the system up and verify V0 span from 20 to 100 V. If O.K., gradually increase Gain by increasing R4: PSU will now work closed loop and self regulate. In this condition very good 100 Hz ripple rejection can be noted. I also tried the 26V raw supply out (in place of 13.6V) and dared to go as high as 140 V, expecting to blow up the 2N3055 at any moment, but this, to my amazement, did not happen. The other figures show the DrDAQ / Picoscope displays. In Picoscope it would be useful to have the RECIPROCAL of the Duty Cycle in MEASUREMENTS.....
5. Breadboard (Medium).jpg
Fig. 10 - The low level circuitry
3.Power unit (Medium).jpg
Fig. 9 - Raw supply and power switching: note no heat sink necessary on the switch
PSU C2 - 100V.jpg
Fig. 8 - Closed Loop display @ V0=100V
PSU C1 - 20V.jpg
Fig. 7 - Closed Loop display @ V0=20V
Loop calculation file
(21 KiB) Downloaded 250 times
Advanced User
Advanced User
Posts: 248
Joined: Sat May 11, 2013 10:03 am
Location: San Marzano Oliveto, Asti, Italy

Return to USB DrDAQ

Who is online

Users browsing this forum: No registered users and 1 guest