Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers
Buy Site Supporter Role (remove some ads) | LPF Donations
Links below open in new window
FrozenGate by Avery
Driver with voltage controlled output
- Thread starter DasGeraet
- Start date
- Joined
- Sep 20, 2013
- Messages
- 17,545
- Points
- 113
No, the diode you linked in the OP can be run using the BB driver with the two batteries you specified there, as well. It has analog modulation and is relatively inexpensive compared to everything you were trying to do at first.
- Joined
- Apr 29, 2008
- Messages
- 1,562
- Points
- 48
Dang nice! I didn't realize there was another IC besides the 5-pin lol. I'll take a look at the schematic tonight and see if I can figure it out.
Last edited:
@paul1598419
You've read that I want to use these cells in parallel, not in series? From what I've read, blackbuck needs minimum 2 cells in series.
@Jimmy
Careful, I'm not entirely sure that it is the TSC101. The Package marking is '101'. There are several current shunt monitors / current sense amplifiers out there with 101 in its name and no other IC than a CSM makes sense in this spot. TSC101 is the only one with the correct lettering on the chip, but the pinout doesn't fit so well. So just think about a CSM and swap the pins a little bit around until it makes sense. By the way, Pin1 not connected is correct, for sure.
You've read that I want to use these cells in parallel, not in series? From what I've read, blackbuck needs minimum 2 cells in series.
@Jimmy
Careful, I'm not entirely sure that it is the TSC101. The Package marking is '101'. There are several current shunt monitors / current sense amplifiers out there with 101 in its name and no other IC than a CSM makes sense in this spot. TSC101 is the only one with the correct lettering on the chip, but the pinout doesn't fit so well. So just think about a CSM and swap the pins a little bit around until it makes sense. By the way, Pin1 not connected is correct, for sure.
Last edited:
- Joined
- Apr 29, 2008
- Messages
- 1,562
- Points
- 48
@badscr
Yeah, I'll probably will do this, and I already designed a board with AD6174 for this purpose. But it's by far not the nicest solution.
Analog modulation of a switchmode supply is not so difficult, even for an aixiz-size driver. I'm just pretty surprised that there is not a single buck-boost out there with this option (or at least with small modifications).
Yeah, I'll probably will do this, and I already designed a board with AD6174 for this purpose. But it's by far not the nicest solution.
Analog modulation of a switchmode supply is not so difficult, even for an aixiz-size driver. I'm just pretty surprised that there is not a single buck-boost out there with this option (or at least with small modifications).
Last edited:
- Joined
- Apr 29, 2008
- Messages
- 1,562
- Points
- 48
Nikolas,
I think there's a few problems with your schematic. TSC101 is a high-side current sense amplifier, and the output is not connected to anything. Also, the FB node has a DC value of 0V, which doesn't make sense.
What bandwidth do you require for your analog modulation? Since the voltage reference is internal, you may have to go with the digipot solution. I'll keep thinking of an alternative.
-Grant
Last edited:
- Joined
- Mar 27, 2011
- Messages
- 514
- Points
- 63
Here let me help:
If that schematic is indeed 100% reverse engineered and nothing is changed, i'm actually interested in the output performance of those TPS63010, because some configuration looks "wrong" to me
also because TI's IC had so much protection scheme it's quite headache making it working with modified regulation scheme like this.
- The SOT-23-5 with marking "101" are ZXCT1010E5TA, you might need to change your schematic accordingly.
- By using high side current sensing which gets its Vcc from the regulator output, you are risking these:
- It will produce overshoot if the regulator output (or whatever Vf you are going to drive) is below 2.5V, because the sense IC needs a minimum of 2.5V to work (see datasheet).
- If your diode's Vf is really below 2.5 at current requirement you said at post #1, then this driver will likely not working (unregulated). Workaround: change to low side current sensing.
- Take caution if you are going to use digital potentiometer to set the gain resistor of sense IC (or opamp) because the bandwidth of digital pot is low, but this TI's IC has 2.2MHz switching frequency. Some regulator needs a certain ripple in its output (which needs to be detected on FB) to regulate its output.
- T1 near the input (which used for battery reverse protection) needs to be able to handle current >2.5A. Or you can actually omit it.
- On some other IC, C1 and C2 might cause oscillation and/or instability. You can actually omit it.
If that schematic is indeed 100% reverse engineered and nothing is changed, i'm actually interested in the output performance of those TPS63010, because some configuration looks "wrong" to me
also because TI's IC had so much protection scheme it's quite headache making it working with modified regulation scheme like this.
@Grant
As i already said, I knew that the TSC101 is not the correct chip. It was something close. Astralist made the correct suggestion, now I'm pretty sure this is the accurate schematic of the Driver. I've updated the schematic in my original post. Could you remove the images of the old schematic from your posts so there is not so much confusion with the 2 different versions here?
I need maximum 1 kHz, so nothing fancy.
@Astralist
Many thanks for the correct suggestion, I've edited the schmatic in my original post. But remember: I reversed this drive, it is my schematic, but not my idea. Dr. Lava designed it and is (or was) selling these since ~2010. It is the only buck/boost I know with variable current output. I think you can read much about the performance of Dr.Lava microflex drive V5 (https://innolasers.com/shop/index.php?id_product=8&controller=product here you can still buy them and there is a datasheet available), I think it even originated in this forum like yours. Never heard of the instabilities suggested by you.
As i already said, I knew that the TSC101 is not the correct chip. It was something close. Astralist made the correct suggestion, now I'm pretty sure this is the accurate schematic of the Driver. I've updated the schematic in my original post. Could you remove the images of the old schematic from your posts so there is not so much confusion with the 2 different versions here?
I need maximum 1 kHz, so nothing fancy.
@Astralist
Many thanks for the correct suggestion, I've edited the schmatic in my original post. But remember: I reversed this drive, it is my schematic, but not my idea. Dr. Lava designed it and is (or was) selling these since ~2010. It is the only buck/boost I know with variable current output. I think you can read much about the performance of Dr.Lava microflex drive V5 (https://innolasers.com/shop/index.php?id_product=8&controller=product here you can still buy them and there is a datasheet available), I think it even originated in this forum like yours. Never heard of the instabilities suggested by you.
Last edited:
- Joined
- Apr 29, 2008
- Messages
- 1,562
- Points
- 48
Hi Nikolas,
I seem to have missed that post, but I see it now.
To modulate the output current, you can inject a current to the FB node. The idea is that whatever current you put into the FB node, the current sense amplifier will put out the opposite to maintain regulation. The ZXCT1010 puts out
I_OUT=((VS+)-(VS-))/100 Ohm,
and with a 50mOhm sense resistor,
(VS+)-(VS-)=I_D*50mOhm,
thus
I_OUT=I_D/2000
So, if you want to increase the output by 1A, then sink 500uA from FB. Now the question is: what bandwidth do you get?
The bandwidth depends on the pulldown resistor on FB, since C1 and C2 will drop your gain at higher frequencies.
F_BW=1/(2*pi*R_PD*440nF)
To get 1kHz bandwidth, you need a pulldown resistance of at most
R_PD=1/(2*pi*1kHz*440nF)=361 Ohms
Uh oh! It looks like in order to get the bandwidth you require, you'll need to either pre-compensate your modulation signal or decrease C1 and C2. Decreasing C1 and C2 to 10nF should be OK, but you'll have to test/simulate to make sure.
Hopefully that gets you in the right direction!
Edit: I may be incorrect about the bandwidth assumption. I'll keep thinking...
Edit2: I believe the bandwidth calculation is correct. If a voltage is injected in the feedback node, the feedback loop will respond according to the loop's bandwidth (assumed to be >1kHz). Since a current is injected, the capacitors will filter out the high frequencies, making a low-passed voltage injection. This voltage is then processed in the feedback loop.
I seem to have missed that post, but I see it now.
To modulate the output current, you can inject a current to the FB node. The idea is that whatever current you put into the FB node, the current sense amplifier will put out the opposite to maintain regulation. The ZXCT1010 puts out
I_OUT=((VS+)-(VS-))/100 Ohm,
and with a 50mOhm sense resistor,
(VS+)-(VS-)=I_D*50mOhm,
thus
I_OUT=I_D/2000
So, if you want to increase the output by 1A, then sink 500uA from FB. Now the question is: what bandwidth do you get?
The bandwidth depends on the pulldown resistor on FB, since C1 and C2 will drop your gain at higher frequencies.
F_BW=1/(2*pi*R_PD*440nF)
To get 1kHz bandwidth, you need a pulldown resistance of at most
R_PD=1/(2*pi*1kHz*440nF)=361 Ohms
Uh oh! It looks like in order to get the bandwidth you require, you'll need to either pre-compensate your modulation signal or decrease C1 and C2. Decreasing C1 and C2 to 10nF should be OK, but you'll have to test/simulate to make sure.
Hopefully that gets you in the right direction!
Edit: I may be incorrect about the bandwidth assumption. I'll keep thinking...
Edit2: I believe the bandwidth calculation is correct. If a voltage is injected in the feedback node, the feedback loop will respond according to the loop's bandwidth (assumed to be >1kHz). Since a current is injected, the capacitors will filter out the high frequencies, making a low-passed voltage injection. This voltage is then processed in the feedback loop.
Last edited:
