开云体育

R&S has some amazing gear but I have avoided it due to unrepairability, lack of service manuals, and lack of support of the older units.?

开云体育

On 7/25/2019 12:58 AM, John kolb wrote:

CPLD's are less complex than FPGA's but not necessarily cheaper as the newer design parts get a lot more logic in a lot less silicon, at the cost of using lower voltages.

Xilinx.? 32 cell = 1.50, 64 cell = 3.25, 128 cell = 7.00, 256 cell = 15 dollars

Spartan 3AN (built in eeprom) 50K gates, about 14 dollars.

Spartan 6, (no eeprom, use winbond = 0.75) about 12 dollars, well over 50 K gates, has a 16 dollar pin compatible version with twice capacity.

Have standardized on the Spartan 6.

Pay careful attention to the bypass capacitors and clock run out to the configuration eeprom.? Programmer available through amazon, web edition (local compile, report back to daddy) is free.

3.3 volts but then again was using them with Xmega and now with ARM.

All chips in non-BGA versions, TQFP-144, so therefore solderable.

SPI interface available through open cores, suggest register structure with read/write registers and go from there.

(and yes, done that....)

Harvey

Lattice 7000S series parts were good for a 5V part, but no longer available new, although still plenty on ebay.

An Altera Max 2 series part (3.3V), mounted on a PC board along with a clock oscillator and 5V to 3.3V voltage regulator would be a good start.
EPM240 series along with a USB Blaster programmer for about $12


Similar modules are available for other CPLDs and FPGAs if the logic doesn't fit the EPM240.

Using a premade module gets one away from having to deal with surface mount parts, at least as long as the interface IC's are still available as thru-hole parts.

I'm still hoping I can figure out how to use the UGPlus to talk to my HP 3456.

John


On 7/24/2019 1:12 PM, Harvey White wrote:

On 7/24/2019 2:01 PM, saipan59 wrote:

I once did a design that had the upD7210 in it (includes controller), driven by an XMEGA.? I've got a similar design (unproven yet) that is driven by an ARM processor.? Since the 75160/75162 are readily available (or at least, were....) I'd stick with them.? I also have a possibility of using an FPGA to do the same.? I'd stick with the FPGA for a new design since it can always be tweaked.? I suspect that a CPLD may be too expensive for the complexity needed at that level.

FPGA goes through level shifters to the 488 bus drivers, so it's all well protected.

After I do a bunch of software (OS/graphics redesign) that's on the list (it says here......)

Harvey

Hi John,

As I said, that was the maximum load instrument. When I first
started working with HPIB, it had just superseded the BCD interface.

My first implementation used 74LS, so my load was about 0.2 inputs,
in its worst case... which were the control lines.

Early instruments would crash the buss if they were turned off, due
to all of the pull down load.

Very soon it became apparent that folks wanted to string their entire
lab's worth of HPIB controlled instruments on their HPIB calculator/
computer, and not worry about whether they were on, or off... So the
lower loss interface drivers had to come out.

The termination resistors are 3K pu, and 6K pd to 5V, so a buss full
of inputs will be at roughly 66% of 5V, or 3.3V... probably a little
higher due to the input leakage currents.

-Chuck Harris

John kolb wrote:

Hello All

I don't often post to the list, but I do read its traffic with interest :-)

It's not a piece of HP test equipment before anyone has a "wet dream".
Its a piece of Rhode & Schwarz test equipment that I rescued from a skip
which was destined to go to landfill.

By rights I shouldn't have done it, but I managed to get permission from
my line manager to take it away as there was nothing classified about
it. That organisations policy was everthing gets crushed and dropped
into a landfill, rather than relieve the bruden on the tax payers and
sell expensive usable test equiment onto small businesses and private
individuals.

It still works, but its MMI has a few bugs and I was thinking of putting
on Ebay (my head says Yes!).
But my heart says I should hold onto it, as its a rare beast.

Best Regards

Danny G7TTR

On 7/25/2019 1:34 AM, John kolb wrote:

The HP 3456 uses MC3448 IC's on all the GPIB lines. This IC is "high impedance" during receive, and open collector and providing pullup conforming to IEEE 448 spec during transmission.

Those are the open collector versions?

I would expect most of the instruments owned by this group to present much less than the 2 TTL loads, so bus loading probably not a problem.

but how many of them?

I would wonder what the bus voltage floats to when EVERYONE is receiving.

Resistive terminations.

Harvey

Haven't dealt with GPIB since about 1970 when I designed a GPIB controlled switch box for internal use at Wavetek. Needed many voltage readings to find the peak values when generating 0.001 Hz with their function generators.

John

On 7/24/2019 3:12 PM, Chuck Harris wrote:

I can answer a couple of your points:

1)? The maximum load any instrument is allowed to present onto the
???? buss is 2 standard TTL inputs.? It was always recommended that
???? the load be only 1 standard TTL input.? Implementations using
???? the TI drivers are much lower than that.

2)? The minimum drive capability (low) of any driver pin is 48ma.

3)? The bus has pull up resistors as well as pull down resistors.
???? I seem to recall 3K pull up, and 6.2K pull down... or maybe it
???? was the other way around...

The numbers work roughly as follows:

1 ttl input load (low) is 1.6ma, 2 is, of course, 3.2ma.

48ma / 3.2ma = 15 maximum, maximum load instruments on the bus at
any one time.

However, the recommendation was only 1 TTL load per input, which would
lead to 30 such instruments total as being the maximum you can drive.

Originally, the intended chips to do the driving were the 7438, which is
a quad NAND OC driver with 48ma per pin capability, and a 7404 as the input.

-Chuck Harris

Reginald Beardsley via Groups.Io wrote:

Hard to know where to begin.

Here is data about the Arduino Uno chip.? It operates from 1.8 to 5.5 V and can source 40 mA from a single pin and source and sink a total of 200 mA for the DIP package.? The SMT package can sink 400 mA.





The board I'm using drives the bus directly. so it probably would not be able to drive a fully populated bus.? The AR488 FW is limited to addresses 1-29 in controller mode.

My knowledge of IEEE-488 is limited to knowing what it is and that the C64 used? a serial implementation for the disk drive interface.? So I have no idea how much drive an instrument requires.? Looking at the SN75160 datasheet it's a bit hard for me to tell whether the ratings are for individual pins or all pins.

I have a total of 12-13 instruments with GPIB interfaces. I'm not sure about the LeCroy DDA-125 and it is much to hard to get to where I could tell by inspection what options it has.

Aside from investigating voltage reference aging and thermal hysteresis, my primary and possibly only application of GPIB would be automating an annual calibration run of my instruments.

I have no need of a cal lab certificate, and as various people have commented,? what cal labs actually do for your money is a bit questionable.? Ideally I'd like to be able to send a single device to a cal lab and then cal everything else to that.? Better still would be to get a traveling set of basic references once a year.

I've generally looked down my nose at the AVR chips, preferring the MSP430 and Stellaris lines from TI and the STM32Fx from ST.? But at $3 for a Chinese copy they look better all the time.

Reg

On 7/25/2019 12:56 AM, Kevin Oconnor wrote:

@Harvey
Good u brought up the upD7210. If that’s the chip I’m thinking of, it is a complete HPIB interface device capable of driving the full complement bus devices.

I think they're barely available.? I've gotten mine out of older instruments that were non functional.? (not HP/Fluke/etc....)? The TMS9914 was only a talker/listener, the 68488 was full.? NI supposedly still sells them, but some companies have been touting their FPGA solution to the problem.? There's an HPIB controller/listener/talker implementation at opencores.org that I've downloaded, and will one day try.

Fluke used it in the later versions GPIB PCBs for the 884X DMMs rather than the discrete design of the earlier boards. One could use the upD7210 with a micro or PI I suppose, and solve the # of units problem. (Is it still available?)

Not sure at all.? I've got a complete CPU board replacement for the Tek DM5010 in the works, which (due to 100 cm * 100 cm cheap Chinese discount board limitations) is built from two boards.? One of them is the 488 bus interface.? The design right now uses a upD7210, but could be reworked to use the original TMS9914 in the DMM itself.? Another option is simply to use the FPGA, which I can program.? Level shifting is on board, and the proper 488 bus drivers are also on board (Tek did use them).

One thing that can be done with that board is to make a single unit properly styled TM5000 plugin that's effectively a smart 488 bus controller with all the custom interfaces, wifi, NRF24L01, etc....? maybe....


Harvey

For me, I have a lot of GPIB units on my benches, and they are all tied to the bus and computer. I don’t want to think about climbing back there to connect up the 2-3 I want to use each time. That’s why the NI device.
Also, working with those old GPIB connectors is a real PITA! Only used cables are affordable, and they are often ratty with flaky or bent screws. If you stack them, you can never predict where they are going to separate! And try to find correct replacement screws, what a pain. (I did find a source recently). And what about the black/silver thread incompatibility! Anyway, I hated that connector in 1975. I still hate it today. Oops, dating myself.
Kevin

Sent from kjo iPhone

开云体育

The HP 3456 uses MC3448 IC's on all the GPIB lines. This IC is "high impedance" during receive, and open collector and providing pullup conforming to IEEE 448 spec during transmission.

I would expect most of the instruments owned by this group to present much less than the 2 TTL loads, so bus loading probably not a problem.

I would wonder what the bus voltage floats to when EVERYONE is receiving.

Haven't dealt with GPIB since about 1970 when I designed a GPIB controlled switch box for internal use at Wavetek. Needed many voltage readings to find the peak values when generating 0.001 Hz with their function generators.

John

CPLD's are less complex than FPGA's but not necessarily cheaper as the newer design parts get a lot more logic in a lot less silicon, at the cost of using lower voltages.

Lattice 7000S series parts were good for a 5V part, but no longer available new, although still plenty on ebay.

An Altera Max 2 series part (3.3V), mounted on a PC board along with a clock oscillator and 5V to 3.3V voltage regulator would be a good start.
EPM240 series along with a USB Blaster programmer for about $12


Similar modules are available for other CPLDs and FPGAs if the logic doesn't fit the EPM240.

Using a premade module gets one away from having to deal with surface mount parts, at least as long as the interface IC's are still available as thru-hole parts.

I'm still hoping I can figure out how to use the UGPlus to talk to my HP 3456.

John

开云体育

Re:-

---

A truly great implementation would use a SBC, and something like TI's
SN75160/161/162BDWR GPIB driver set... which are still available, and do
have the capability to drive a full buss... Mouser has them, in stock.

But, alas, you won't be driving that chip set from most USB ports.

-Chuck Harris

---

Odd, because that's exactly what NI, Agilent/Keysight, Keithley do with their USB/GPIB adapters, that are powered from the host PC over the USB port.

Also..

As most, if not all, of the Arduino or Pic based USB/Serial to GPIB adapters are open source, code and schematic, it's relatively simple to add "proper" bus drivers if you needed.

The difficult bit, will be negotiating with the host PC, for it to allow enough current to be pulled from the USB port to power it all, if a "Full" GPIB system is ever connected...

No one, and I mean no one who knows SBC's like the Pi series or similar, would even consider connecting a GPIB bus to the GPIO pins, for one thing, GPIB is 5V, and the GPIO pins are 3V logic, some may be 5V tolerant, but even then that's not good practice.?? But there is no reason not to put some logic between it and some bus drivers, that use the full 5V to run the bus.

Many years back, I managed to crudely talk to one simple instrument (a MI power meter) by bit-banging the LPT port on a Win9x system to emulate a GPIB controler.?? Slow and fragile, but it worked.? (No, I don't have that code, or notes any more, it was a "just because I can" experiment that got a somewhat out of hand...)

No drivers were harmed.

Regards.

Dave B

-- 
Created on and sent from a Unix like PC running and using free and open source software:

@Harvey
Good u brought up the upD7210. If that’s the chip I’m thinking of, it is a complete HPIB interface device capable of driving the full complement bus devices. Fluke used it in the later versions GPIB PCBs for the 884X DMMs rather than the discrete design of the earlier boards. One could use the upD7210 with a micro or PI I suppose, and solve the # of units problem. (Is it still available?)
For me, I have a lot of GPIB units on my benches, and they are all tied to the bus and computer. I don’t want to think about climbing back there to connect up the 2-3 I want to use each time. That’s why the NI device.
Also, working with those old GPIB connectors is a real PITA! Only used cables are affordable, and they are often ratty with flaky or bent screws. If you stack them, you can never predict where they are going to separate! And try to find correct replacement screws, what a pain. (I did find a source recently). And what about the black/silver thread incompatibility! Anyway, I hated that connector in 1975. I still hate it today. Oops, dating myself.
Kevin

Sent from kjo iPhone

On Wed, Jul 24, 2019 at 04:19 PM, saipan59 wrote:

* On Wed, Jul 24, 2019 at 02:46 PM, jafinch78 . wrote:

I've wondered the same for SCSI hard drive adapters to be able to upgrade
older equipment to IDE/PATA/SATA hard drives.

No specific example, but I'm pretty sure that sort of thing has been done
multiple times. As you can surmise, there is a great need among 'vintage
computer' collectors/users. Perhaps not as much for SCSI-to-modern (yet), but
definitely for pre-SCSI. Years ago I made a couple of interfaces for DEC
PDP-11 systems to replace a floppy drive with a modern laptop (did the floppy
because the protocols are simple).

Pete

Circa 1990 a Sun SCSI shoebox consisted of an ESDI drive and an Emulex SCSI - ESDI adapter card. After I got a 1.8 GB SCSI drive, I traded two 140 MB ESDI drives for a wonky Tek 465. Which worked *just* well enough to allow me to fix my Dumont 1060 horizontal sweep which then allowed me to fix the various faults in the 465. At which point I felt entitled to say I had some ability in electronics rather than being a mere dabbler. It took all my free time for several weeks. But even 30 years later I'm proud of doing that with just the manuals and no one to help me.

Hi all,

I have a 431B power meter that I acquired a while back for very cheap. Now that I not only have more space to work, and have a working thermistor mount that came with the 432A power meter I use mostly, I thought I'd sit down and see about fixing the 431B.

It has the following symptoms: when powered on? in the null position of the range switch, the needle floats above the null region and the potato slicer variable cap that controls null has no effect. Additionally, the zeroing knob and vernier do nothing. If the unit is switched to any position other than null, the meter pegs hard full scale.

I have gone through the listed checks in the manual, and have found that all the dc voltages for the power supply, the 10 KHz oscillator-amplifier, and 10 KHz amplifier are nominal. Ripple is in spec. However, it all goes wrong in the squaring circuit, and this is where I'm getting confused. The cathodes of CR109-CR113 are supposed to have a positive dc voltage, with CR108 sitting at 0 Vdc (listed in a Table 5-7 of the manual), but all of mine are around -2.6 Vdc. These voltages are supposed to be set by the voltage dividers formed by the adjacent 1/2% resistors, none of which look damaged in any way.

I haven't tried pulling legs of components out for testing yet, but I was wondering if anyone has seen this problem before?

Thanks!

Sean

On 2019-07-24 4:46 p.m., jafinch78 . wrote:

On Wed, Jul 24, 2019 at 10:38 AM, David Holland wrote:

How's come nobody has done a Raspberry PI GPIB "Hat", with a real
driver interface chips?

Is there a lack of I/O pins, or power?

I've wondered the same for SCSI hard drive adapters to be able to
upgrade older equipment to IDE/PATA/SATA hard drives.?

Has anyone designed an open source MCU/SBC based SCSI adapter for modern
hard drives that is open source and doesn't require a translator?

The SCSI2SD is reported to work well.




--Toby

- James Finch

?

I can answer a couple of your points:

1) The maximum load any instrument is allowed to present onto the
buss is 2 standard TTL inputs. It was always recommended that
the load be only 1 standard TTL input. Implementations using
the TI drivers are much lower than that.

2) The minimum drive capability (low) of any driver pin is 48ma.

3) The bus has pull up resistors as well as pull down resistors.
I seem to recall 3K pull up, and 6.2K pull down... or maybe it
was the other way around...

The numbers work roughly as follows:

1 ttl input load (low) is 1.6ma, 2 is, of course, 3.2ma.

48ma / 3.2ma = 15 maximum, maximum load instruments on the bus at
any one time.

However, the recommendation was only 1 TTL load per input, which would
lead to 30 such instruments total as being the maximum you can drive.

Originally, the intended chips to do the driving were the 7438, which is
a quad NAND OC driver with 48ma per pin capability, and a 7404 as the input.

-Chuck Harris

Reginald Beardsley via Groups.Io wrote:

开云体育

On Wed, Jul 24, 2019 at 06:05 AM, Chuck Harris wrote:

These days, microprocessor cores are combined with high current
tristate/open collector drivers, clock oscillators, Uarts, ADC's
and DACS, RAM/ROM/FlashRAM, ... kitchen sink... all on a single
monolythic chunk of silicon.

Some of these wonderful devices, such as the PIC family can source
and sink 25ma from each I/O pin... as long as the total from the
chip doesn't exceed something like 300ma.

Hi Chuck, thanks, and yes I'm very familiar with the modern MCU situation. My off-hand remark yesterday was based on a no-data assumption that most modern MCU pins wouldn't be able to meet GPIB electrical specs. I understand that most major MCU vendors still have some 5V parts which may have fairly beefy output drivers, but those parts are generally fading away - not much used in new designs. Most of what I've used in the last 10 years or so for projects are 3.3V parts with 'maybe' 4 mA of drive strength.
It's interesting that the old 5V parts are 'good enough' for at least a basic implementation that works (without really meeting the specs).

It's one of those silicon tradeoffs.? The Xmega/Mega series of processors are designed with higher capability were supposed to be more standalone than the larger ARM processors.? I suspect that looking at the smaller ARM processors would have some of those higher capability drive pins.? MEGA would run from 3.3 volts or 5.0? volts (3.3 at a reduced clock speed).? XMEGA did 3.3 volts only, higher clock speed, etc.? Different markets, I'll bet.

Harvey

Pete

• On Wed, Jul 24, 2019 at 02:46 PM, jafinch78 . wrote:

I've wondered the same for SCSI hard drive adapters to be able to upgrade older equipment to IDE/PATA/SATA hard drives.?

No specific example, but I'm pretty sure that sort of thing has been done multiple times. As you can surmise, there is a great need among 'vintage computer' collectors/users. Perhaps not as much for SCSI-to-modern (yet), but definitely for pre-SCSI. Years ago I made a couple of interfaces for DEC PDP-11 systems to replace a floppy drive with a modern laptop (did the floppy because the protocols are simple).

Pete

Hard to know where to begin.

Here is data about the Arduino Uno chip. It operates from 1.8 to 5.5 V and can source 40 mA from a single pin and source and sink a total of 200 mA for the DIP package. The SMT package can sink 400 mA.





The board I'm using drives the bus directly. so it probably would not be able to drive a fully populated bus. The AR488 FW is limited to addresses 1-29 in controller mode.

My knowledge of IEEE-488 is limited to knowing what it is and that the C64 used a serial implementation for the disk drive interface. So I have no idea how much drive an instrument requires. Looking at the SN75160 datasheet it's a bit hard for me to tell whether the ratings are for individual pins or all pins.

I have a total of 12-13 instruments with GPIB interfaces. I'm not sure about the LeCroy DDA-125 and it is much to hard to get to where I could tell by inspection what options it has.

Aside from investigating voltage reference aging and thermal hysteresis, my primary and possibly only application of GPIB would be automating an annual calibration run of my instruments.

I have no need of a cal lab certificate, and as various people have commented, what cal labs actually do for your money is a bit questionable. Ideally I'd like to be able to send a single device to a cal lab and then cal everything else to that. Better still would be to get a traveling set of basic references once a year.

I've generally looked down my nose at the AVR chips, preferring the MSP430 and Stellaris lines from TI and the STM32Fx from ST. But at $3 for a Chinese copy they look better all the time.

Reg

On Wed, Jul 24, 2019 at 10:38 AM, David Holland wrote:

How's come nobody has done a Raspberry PI GPIB "Hat", with a real
driver interface chips?

Is there a lack of I/O pins, or power?

I've wondered the same for SCSI hard drive adapters to be able to upgrade older equipment to IDE/PATA/SATA hard drives.?

Has anyone designed an open source MCU/SBC based SCSI adapter for modern hard drives that is open source and doesn't require a translator?

- James Finch

?