50V Drain Pulser Board artwork

Hi DavangP,

                         Attached is the 50 V pulser artwork you requested.  

I also would like to share a new 50 V pulser we have recently released, It is called Pulser Plus, it is fully compatible with the ADPA1106. The advantages of the pulser pulse are, on board Vneg generation for the gate bias, a resistor is used to set the desired bias point for the amplifier and is capable of drain pulsing or gate pulsing. The documentation and design files can be found here:  Pulser Plus Reference Design for GaN PA Biasing and Sequencing [Analog Devices Wiki]

Regards,

Jim Bedrosian

7331.50V_ Pulser_09-065340-01C.zip

Thanks Jim....I read the documentation for the Pulser Plus board (specifically the schematic) and its recommended operating conditions are < 300us Pulse width and < 20% duty cycle.  If I had to operating a part greater than 300us pulse width and/or 20% duty, do I need to modify the pulser board to support those conditions?  Thank you.

Hi DavangP,

In itself the core of the circuit doesn’t really care much about PW, duty-cycle or time-period. As example, the drain can remain fully on all the time. It is actually the case when running Gate Pulsing when you think about it. As for Gate Pulsing, similar comment could be said. The main limitation comes from the power dissipation of the power amp. The maximum pulse-width and period were a requirement defined in the design process.

Now, this being said, there are a few sections in the circuit that must be looked at when it comes to operating at much higher duty-cycles / time periods. I “tagged” them with a double-asterisk in schematic. Resistors R28, R3, R13, R54 and R55. The dissipation for these specific resistors can be important when operating at high duty-cycles. I didn’t want to use physically large resistors for those locations to handle the much higher dissipation that’d come with the wider duty-cycle / longer time periods. So, they are rated for high power surges but would be limited the closer you get to DC. The resistors:

• R28 (For Drain voltage monitoring when using 50ohm termination): Could change to a higher resistor value. Attenuation would increase but power dissipation would decrease. Either that or use a 1Meg termination. For the rise/fall time measurements with 50hm term, a shorter pulse could be considered
• R3 (emulate PA load for testing Drain pulse without an amp): For electrical tests only, so shouldn’t be necessary anymore. But if desired, 10ohm is for 5A at 50V. Could size it up according to the amplifier that’s going to be connected. That’d help. Otherwise, could consider adding a resistor with a much bigger package across VDD_PA and GND.
• R13 (Drain_Pulse_Enable termination): Not needed for gate pulsing. For drain pulsing, a 200ohm could be used. With a 5V input signal, that’s 250mW dissipation DC, so right at rating. Obviously, it can be adjusted to match pulse-width/duty-cycle along with input signal amplitude. So, somewhere between 50ohm and 200ohm could be considered if termination is really desired.
• R55: leave it DNI
• R54 (gate drive resistance): That one is trickier. It is based on a certain amount of gate capacitance and certain rise/fall times rate. The duty-cycles shouldn’t help much here since the number of rise/fall times remain the same. If, however, the time period is much longer then it should help. So, on paper, it should all be good. But worth keeping an eye for it.

 Regards,

Jim B

Hi Jim, In the Pulser Plus Board, You have mentioned the maximum drain current that the board can support as 5A. What is the limiting factor of the load current in this board? Can i connect a PA of 10A load to this Pulser Plus Board?

Also, you are saying that R3 can be used as a load with the operating conditions of 50V, 5A. 
The wattage of the R3 in the schematic is 1.5W. But according to the operating conditions, the wattage on the resistor would be 250W which is very high and obiously the resisttor would burn. Thus, how can we operate the board in this condition?

HI Utsech,

In regards to R3. Some added details:

• The part used is the CRCW251210R0FK**HP  (** is for the package size).

• Datasheet: https://www.vishay.com/docs/20043/crcwhpe3.pdf
• DC/average power dissipation is 1.5W at 70C ambient per datasheet:

Dissipation decreases at higher ambient temps:

• When it comes to short duration pulses, the part can tolerate significantly higher powers. The CRCW-HP pulsed rated series was selected specifically for that reason. Many resistors can tolerate much more power as long as a certain duration isn’t exceeded. In case of the Vishay -HP series, the pulsed power are tested so it is relatively easy to make sure if the part selected can tolerate the power levels. For the 10ohm, at 50V (5A), we’re talking 250W worst-case. We specified a maximum pulse width of 300us. So, if I am to look at the graphs in the CRCW-HP series datasheet, I get

Around 400-500W for 300us (single pulse)

Periodic/Continuous pulses would lower that value.

However, since the point of the resistor is to verify the electrical functionality of the driver, a single pulse should be good enough. If continuous pulses were to be used, a really short duty-cycle (like 1 pulse every minute) could be considered.

• Although R3 is only present for electrical verification as I just mentioned. The real verification can only be done in combination with the use case PA.

Regards,

Jim B

Hi Utsech,

About the current limit. The LTC7000A does provide plenty of information. I’d suggest starting by reading it in more details:

• To adjust the current limit:
  • Iset PIN section: https://www.analog.com/media/en/technical-documentation/data-sheets/ltc7000a.pdf#page=15
  • The current is dVth/Rsns. So, if Rset is equal to 120kOhm, that means dVth = 60mV. ECB current would be 60mV/Rsense. So, if a 10mOhm Rsense were to be used, the nominal threshold for the ECB would be 6A.
• The MOSFET may have to be modified depending on the application requrieemtns. The MOSFET selection section is a good starting point ( https://www.analog.com/media/en/technical-documentation/data-sheets/ltc7000a.pdf#page=20  

Regards,

Jim B

Thank you for the response Jim. I am testing the pulserplus board. I am supplying 28V.
The gate pulse circuitry is working but the drain pulse is not coming.

When i am providing the drain pulse enable, the drain output is zero with or without connecting the load at the drain output. 
I am getting the pulsed signal on the Imon pin from 0V to 1.6V when i just gave the supply.
What may be the reason for this behaviour on the Imon pin. As per the datasheet if Imon pin is above 1.5V, then there must be some overload. But how can the Imon pin detect 1.6V even though i am not connecting any load.

The Resistance value near the Iset pin is 100K. The resistance value near the Imon pin is 20K.

I want to obtain a drain pulse of 28V and 50V.

Please suggest if I have to change any resistance or capacitor values to obtain my desired output.

Hi Utstech,

Here’s what I’d recommend: 

1. With a multimeter, measure the resistance between the VDD_PA port and GND. To make sure parts like Q2 or D1 have not been damaged.
2. Using only the PULSERPLUS board (no amp board attached) and using short pulses (not more than 50us) applied to the DRAIN_PULSE_ENABLE connector, differentially measure the voltage between the I_SENSE_MON_P and I_SENSE_MON_N pins. If the LTC7000A reaches current limit, the voltage seen should toggle between 0V and >600mV. If the LTC7000A detects and overcurrent, it’ll turn off the FET Q1 within micro-seconds. Faster than the 50us pulses applied on the DRAIN_PULSE_ENABLE. That’d confirm the overcurrent hypothesis.

Regards,

Jim B

Hi Jim,
I am still observing the tripping. The VDD_PA output pulse is not turning ON completely. The pulse is turning OFF before turning ON fully. This is happening as soon as i turn ON the power supply to the board. 

The signal is fluctuating between overload(1.65V which is above 1.5V) and 0V at the Imon pin. 

**I am supplying 28V supply and connecting 16ohm load across the VDD_PA and ground.