encoder voltage
- Gorts Revenge
- Offline
- Junior Member
Less
More
- Posts: 23
- Thank you received: 0
01 Jan 2023 07:23 #260701
by Gorts Revenge
encoder voltage was created by Gorts Revenge
Can the 7i76E handle a ~25v differential encoder input?
If not, does anyone have any clever work-around solutions?
My AC servo drive has a ~25V differential encoder output that I was hoping to use.
If not, does anyone have any clever work-around solutions?
My AC servo drive has a ~25V differential encoder output that I was hoping to use.
Please Log in or Create an account to join the conversation.
- tommylight
- Away
- Moderator
Less
More
- Posts: 19464
- Thank you received: 6529
01 Jan 2023 11:13 #260702
by tommylight
Replied by tommylight on topic encoder voltage
Google voltage divider, you would need two for each encoder output/input.
Also using 4.7V zener diodes in parallel with inputs is a good idea.
Also using 4.7V zener diodes in parallel with inputs is a good idea.
Please Log in or Create an account to join the conversation.
- Gorts Revenge
- Offline
- Junior Member
Less
More
- Posts: 23
- Thank you received: 0
01 Jan 2023 17:03 - 01 Jan 2023 17:11 #260714
by Gorts Revenge
Replied by Gorts Revenge on topic encoder voltage
Thanks for the quick reply. I'm sure that I'll be preaching to the choir here but there may be some other readers that aren't in the choir so it's probably worth the back and forth. I'm not trying to be belligerent.
A simple voltage divider circuit is a reasonable suggestion but it prompts some other questions. For those that don't know, it's just a pair of resistors between the source and a ground with the circuit's output being the node between them. Zeners are special diodes that avalanche in reverse when a calibrated voltage limit has been reached (they're like an electric blow-off valve) so, In this case, they're being suggested as a safety/clean-up measure. The voltage divider circuit can work well depending on the current requirements of the load and the turn-down requirements; however, the entropy of the circuit will stack rapidly relative to the voltage turn-down. If this particular circuits' current requirement is more than a couple of miliamps, it will become unacceptably high, resulting in an excessive load at the source. In other words, if V1 is 25V and V2 needs to be under 5V, a resistor of less than ~ 10K will eventually kill the drive's encoder OP circuit. The smaller the resistance, the faster it will die. A 10K will probably provide enough push for the Mesa but more information is needed. Using a network that equals 12.5K should draw roughly 50 mW and push about 2 mA. Is the 2mA sufficient? Is the 50mW too much? Usually, these circuits are happy under 10 mW.
Apart from a stand-alone divider network, I was also considering using smaller divider networks to push some VN2222 fets or some 18V hex buffers but what I was really hoping for was a miracle cure (plug-in, finished product solution). Making circuit boards is like going to the dentist. I'll post a schematic of whatever I come up with.
Thanks again.
A simple voltage divider circuit is a reasonable suggestion but it prompts some other questions. For those that don't know, it's just a pair of resistors between the source and a ground with the circuit's output being the node between them. Zeners are special diodes that avalanche in reverse when a calibrated voltage limit has been reached (they're like an electric blow-off valve) so, In this case, they're being suggested as a safety/clean-up measure. The voltage divider circuit can work well depending on the current requirements of the load and the turn-down requirements; however, the entropy of the circuit will stack rapidly relative to the voltage turn-down. If this particular circuits' current requirement is more than a couple of miliamps, it will become unacceptably high, resulting in an excessive load at the source. In other words, if V1 is 25V and V2 needs to be under 5V, a resistor of less than ~ 10K will eventually kill the drive's encoder OP circuit. The smaller the resistance, the faster it will die. A 10K will probably provide enough push for the Mesa but more information is needed. Using a network that equals 12.5K should draw roughly 50 mW and push about 2 mA. Is the 2mA sufficient? Is the 50mW too much? Usually, these circuits are happy under 10 mW.
Apart from a stand-alone divider network, I was also considering using smaller divider networks to push some VN2222 fets or some 18V hex buffers but what I was really hoping for was a miracle cure (plug-in, finished product solution). Making circuit boards is like going to the dentist. I'll post a schematic of whatever I come up with.
Thanks again.
Last edit: 01 Jan 2023 17:11 by Gorts Revenge. Reason: typos
Please Log in or Create an account to join the conversation.
- blazini36
- Offline
- Platinum Member
Less
More
- Posts: 927
- Thank you received: 143
01 Jan 2023 19:50 #260725
by blazini36
Replied by blazini36 on topic encoder voltage
24v differential encoder output seems odd. Encoder outputs generally want to follow RS422 specs and should be 5v. You didn't mention the resolution, speed or anything so you can't say what frequency this thing will be operating at which is a major factor.
I use UDN2803's as a general purpose FieldV input -> MCU since it's pretty fast but it's output is inverted which shouldn't matter a bit for encoders. I tend to use a MIC2891/2892 for outputs but the chip's inputs can handle 30v so it can be used for field-logic inputs as well. The MIC2891 is a little slower than the UDN2803 but it's non-inverting. But both are probably fine depending on the frequencies you're looking at. There's probably a bunch of ICs I'm unaware of that would be fine but for the $3 a multi input IC costs it's pointless to much around with discrete passives.
I use UDN2803's as a general purpose FieldV input -> MCU since it's pretty fast but it's output is inverted which shouldn't matter a bit for encoders. I tend to use a MIC2891/2892 for outputs but the chip's inputs can handle 30v so it can be used for field-logic inputs as well. The MIC2891 is a little slower than the UDN2803 but it's non-inverting. But both are probably fine depending on the frequencies you're looking at. There's probably a bunch of ICs I'm unaware of that would be fine but for the $3 a multi input IC costs it's pointless to much around with discrete passives.
The following user(s) said Thank You: Gorts Revenge
Please Log in or Create an account to join the conversation.
Time to create page: 0.053 seconds