Re: [Dorset] CV148 Description and Spec
Hi Terry On 06/06/18 15:44, Terry Coles wrote: I'm not sure that I understand that. My 1-Wire Temperature Sensors (https:// www.amazon.co.uk/gp/product/B00HCB8GLU/ref=oh_aui_detailpage_o01_s00? ie=UTF8=1) only have three wires; +3.3 V, 0 V and data. When I connect them to my Pi I can read the temperature without any messing with pulse generation or measurement. That's fine - whoever has written the 1-Wire interface driver software will have needed to mess with those things. I can see that if we wanted to design the interface in discrete components or low-level logic, we would probably need to think about such things, but even we aren't daft enough to try that :-) What we need is the 1-Wire interfacing components that are integrated into the DS18B20. I wasn't suggesting that, only implementing the 1-Wire driver code on the Pi, but I gather that's all been done. In terms of multiplexing several Hall effect sensors, you could use a PISO shift register (eg 74HC165) and clock the bits out as with SPI. For longer distances, I'd suggest a single chip micro to read the Hall devices and send the data over asynchronous serial (RS232, RS485 etc). An MC9S08SH8 is available in DIL package and a USBDM programmer can be bought on Ebay for around £20. We are happy with getting the bits into serial and have considered several designs. We've got to the point of deciding the best interfacing technique at the moment. Currently we are prototyping a system using I/O expander chips and I2C. If that throws up problems then we'll look at SPI, 1-Wire or RS232/422/485 and whatever logic we need to get the bits to the bus. Looks like Ralph has located what you were looking for! Cheers Tim -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
Re: [Dorset] CV148 Description and Spec
On Wednesday, 6 June 2018 16:06:28 BST Ralph Corderoy wrote: > > Yes, but I've yet to find a 1-wire I/O expander. > > https://www.maximintegrated.com/en/products/digital/memory-products/DS2408.h > tml https://datasheets.maximintegrated.com/en/ds/DS2408.pdf Ralph, Do you have a degree in Googling? I searched for a device like this and found nothing useful. Thanks. -- Terry Coles -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
Re: [Dorset] CV148 Description and Spec
Hi Terry, > Yes, but I've yet to find a 1-wire I/O expander. https://www.maximintegrated.com/en/products/digital/memory-products/DS2408.html https://datasheets.maximintegrated.com/en/ds/DS2408.pdf Cheers, Ralph. -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
Re: [Dorset] CV148 Description and Spec
On Wednesday, 6 June 2018 15:10:10 BST t...@ls83.eclipse.co.uk wrote: > With I2C you are typically limited to just a small number of identical > devices on the same bus, as they come preprogrammed with a 7-bit > identifier, with the ability to change perhaps 1 or 2 of those bits (but > if there are no spare pins available on the device, it won't be possible Understood, but the I/O expander chips are (literally) as cheap as chips, so we will have one per sensor. We only need around three or four devices on each Pi anyway. > to select any address bits). 1-Wire devices come with a 64-bit > identification code (8 bit family code, 48 bit serial number, 8 bit > CRC). This allows multiple identical devices to hang off the same bus > and be uniquely identified. Identification of all devices on the bus is > via a search routine implemented in the master. Yes, but I've yet to find a 1-wire I/O expander. > One problem with 1-Wire is that it requires time-critical pulse > generation and measurement. The DS2282-100 IIC to 1-Wire bridge allows > you to avoid time-critical stuff on the master processor. I'm not sure that I understand that. My 1-Wire Temperature Sensors (https:// www.amazon.co.uk/gp/product/B00HCB8GLU/ref=oh_aui_detailpage_o01_s00? ie=UTF8=1) only have three wires; +3.3 V, 0 V and data. When I connect them to my Pi I can read the temperature without any messing with pulse generation or measurement. I can see that if we wanted to design the interface in discrete components or low-level logic, we would probably need to think about such things, but even we aren't daft enough to try that :-) What we need is the 1-Wire interfacing components that are integrated into the DS18B20. > In terms of multiplexing several Hall effect sensors, you could use a > PISO shift register (eg 74HC165) and clock the bits out as with SPI. > For longer distances, I'd suggest a single chip micro to read the Hall > devices and send the data over asynchronous serial (RS232, RS485 etc). > An MC9S08SH8 is available in DIL package and a USBDM programmer can be > bought on Ebay for around £20. We are happy with getting the bits into serial and have considered several designs. We've got to the point of deciding the best interfacing technique at the moment. Currently we are prototyping a system using I/O expander chips and I2C. If that throws up problems then we'll look at SPI, 1-Wire or RS232/422/485 and whatever logic we need to get the bits to the bus. -- Terry Coles -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
Re: [Dorset] CV148 Description and Spec
Hi Terry On 06/06/18 13:53, Terry Coles wrote: Pros and Cons of I2C and 1-wire Interfaces: This is to communicate with multiple devices connected to a Raspberry Pi. Two fundamental elements are relevant. I2C (literally Inter-Integrated Circuit) is only good for short distances, but has the advantage of allowing addressing as part of the protocol and sensor design. 1-Wire has longer range, but the devices arrive with their unique identifier hard coded into them by the manufacturer. After last night, I was unsure whether we would be able to implement a level sensor with 1-wire capability, since they tend to be very expensive and aimed at specific industries and therefore expensive. Our current level sensors are using hall-effect devices with a bit per level, so we clearly want to multiplex that data into a serial bus of some kind. This appears fairly easy with I2C, but not so easy (because of the unique ID requirement) with 1-wire. The jury's still out. With I2C you are typically limited to just a small number of identical devices on the same bus, as they come preprogrammed with a 7-bit identifier, with the ability to change perhaps 1 or 2 of those bits (but if there are no spare pins available on the device, it won't be possible to select any address bits). 1-Wire devices come with a 64-bit identification code (8 bit family code, 48 bit serial number, 8 bit CRC). This allows multiple identical devices to hang off the same bus and be uniquely identified. Identification of all devices on the bus is via a search routine implemented in the master. One problem with 1-Wire is that it requires time-critical pulse generation and measurement. The DS2282-100 IIC to 1-Wire bridge allows you to avoid time-critical stuff on the master processor. In terms of multiplexing several Hall effect sensors, you could use a PISO shift register (eg 74HC165) and clock the bits out as with SPI. For longer distances, I'd suggest a single chip micro to read the Hall devices and send the data over asynchronous serial (RS232, RS485 etc). An MC9S08SH8 is available in DIL package and a USBDM programmer can be bought on Ebay for around £20. Cheers Tim -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
Re: [Dorset] CV148 Description and Spec
On Wednesday, 6 June 2018 12:18:43 BST Ralph Corderoy wrote: > A couple more things from last night. > > In 1942, silver was available when copper was in short supply so > $1,000,000,000, in 1942's terms, was used to build a centrifuge. They > went to some effort to retrieve every scrap afterwards; detail is on > Wikipedia. Also, we discussed a couple of other topics on our side of the table: URL Fitering: How to filter URLs using a firewall running on a Raspberry Pi that happens to be also running the webserver. (This is the Android "This WiFI Access point has no Internet access" problem again, which morphs every time Google bring out a new version of Android). Paul talked me through it and provided a couple of links: https://duckduckgo.com/?q=url+filtering+raspberry+pi=canonical=web https://computers.tutsplus.com/articles/installing-openwrt-on-a-raspberry-pi-as-a-new-home-firewall--mac-55984 and - squidguard on pi: http://danscourses.com/turn-a-raspberry-pi-into-a-web-filter-proxy-with-squidguard/ Pros and Cons of I2C and 1-wire Interfaces: This is to communicate with multiple devices connected to a Raspberry Pi. Two fundamental elements are relevant. I2C (literally Inter-Integrated Circuit) is only good for short distances, but has the advantage of allowing addressing as part of the protocol and sensor design. 1-Wire has longer range, but the devices arrive with their unique identifier hard coded into them by the manufacturer. After last night, I was unsure whether we would be able to implement a level sensor with 1-wire capability, since they tend to be very expensive and aimed at specific industries and therefore expensive. Our current level sensors are using hall-effect devices with a bit per level, so we clearly want to multiplex that data into a serial bus of some kind. This appears fairly easy with I2C, but not so easy (because of the unique ID requirement) with 1-wire. The jury's still out. -- Terry Coles -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
Re: [Dorset] CV148 Description and Spec
Hi Terry, > Peter has brought a bit of history to the Meeting; a CV148 infrared > image converter, dating from the late 40s. A couple more things from last night. In 1942, silver was available when copper was in short supply so $1,000,000,000, in 1942's terms, was used to build a centrifuge. They went to some effort to retrieve every scrap afterwards; detail is on Wikipedia. "How much do you need?" I replied, "Six thousand tons." 'How many troy ounces is that?" he asked. — https://en.wikipedia.org/wiki/Calutron#Construction Ivan Sutherland's 1988 Turing-Award paper on `Micropipelines' used for asynchronous linkage of blocks in a circuit instead of a synchronous clock. https://dl.acm.org/citation.cfm?id=63532 Free PDF download. Cheers, Ralph. -- Next meeting: Bournemouth, Tuesday, 2018-07-03 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
[Dorset] CV148 Description and Spec
Hi, Peter has brought a bit of history to the Meeting; a CV148 infrared image converter, dating from the late 40s. Here is the link to the description and spec: http://www.tubecollector.org/documents/cv14x.htm -- Terry Coles -- Next meeting: Bournemouth, Tuesday, 2018-06-05 20:00 Meets, Mailing list, IRC, LinkedIn, ... http://dorset.lug.org.uk/ New thread: mailto:dorset@mailman.lug.org.uk / CHECK IF YOU'RE REPLYING Reporting bugs well: http://goo.gl/4Xue / TO THE LIST OR THE AUTHOR
