Mike Monett wrote:
> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
> [...]
>>> I shall write them and ask where they get those positive pulses.
>>
>> Probably from a TIA.
>>
>> One approach is to use an old-fashioned low-gain MMIC--most are
>> basically Darlingtons with resistive feedback, so they invert.
>>
>> Cheers
>>
>> Phil Hobbs
>
> I begin to suspect a simple inverter. This shows the schematic of the
> Theremino about 1/3 of the way down. 100 Meg feedback to a flyback HV
> generator, 2 Meg PMT load, grounded cathode, pulse-shaping network to a
> grounded emitter inverter. No concern about linearity. One thing I don't
> understand is a 1N4148 connected betwen the base and ground to catch
> negative pulses. I wonder if this is a cheap log converter:
>
> https://physicsopenlab.org/2016/01/26/diy-gamma-spectrometry/
>
> I think I'll run mine with a negative cathode to avoid mistakes wiping out
> the preamplifier. It's not going to be in operation long enough to worry
> about ion migration. Then find out how RH inverts their pulses.
>
>

Looks like it's for short circuit protection. A spark across the output
terminals could be quite unpleasant for T3 otherwise.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

[...]

> Looks like it's for short circuit protection. A spark across the output
> terminals could be quite unpleasant for T3 otherwise.
>
> Cheers
>
> Phil Hobbs

That is precisely why I do not want to run mine with 800 Volts on the anode.

--
MRM

On Fri, 22 Jul 2022 12:30:37 +0300, Dimiter_Popoff <dp@tgi-sci.com>
wrote:

>On 7/22/2022 3:17, John Larkin wrote:
>> On Thu, 21 Jul 2022 23:53:53 -0000 (UTC), Mike Monett <spamme@not.com>
>> wrote:
>>
>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>
>>>> Mike Monett wrote:
>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>
>>>>> [...]
>>>>>
>>>>>> Yup. Back in the long ago (1984ish), I built a power supply filter
>>>>>> that got rid of 10 Vpp of 120 Hz ripple on a 2 kV supply for some
>>>>>> piezo stacks. It worked by lifting the cold end, sensing the hot end
>>>>>> via a 100 nF, 3 kV film cap, and wiggling the cold end to keep it
>>>>>> still. (It had some TVS protection too, obviously.)
>>>>>>
>>>>>> With an LF356 op amp, I got 100 dB of ripple rejection, which came in
>>>>>> pretty handy. I could probably have done considerably better with
>>>>>> that feedforward trick of Woodward's, where you use another op amp to
>>>>>> measure the error voltage of the main one, and add in its output.
>>>>>>
>>>>>> Another approach like that is the Kanner Kap, which wasn't first
>>>>>> invented by Kanner, but he now owns it on account of the cute name.
>>>>>> That's the trick where you use an RC lowpass, and drive the cold end
>>>>>> of the cap with some super beefy amplifier to make the top stay still.
>>>>>>
>>>>>> Cheers
>>>>>>
>>>>>> Phil Hobbs
>>>>>
>>>>> The CCFL inverter I got from Amazon runs at 40KHz.
>>>>>
>>>>> A simple half-wave rectifier running at 40 KHz with a 1nF cap and 9e6
>>>>> Ohm load will produce about 2V sawtooth ripple.
>>>>>
>>>>> This is about 2/800 = 0.0025 * 100 = 0.25% ripple.
>>>>>
>>>>> That should be good enough.
>>>>
>>>> Totally. You can probably use one of the HV caps that came with it to
>>>> couple the anode pulses to the outside world.
>>>>
>>>> Cheers
>>>>
>>>> Phil Hobbs
>>>
>>> That is a problem. The RH Electronics MCA expects positive pulses. These
>>> won't come from the anode.
>>>
>>> https://static.wixstatic.com/media/e43988_a63520f4ed07436a843ddc1be0fda46a~
>>> mv2.jpg
>>>
>>> https://video.wixstatic.com/video/e43988_8006efb319884c72b9d7f7418abdfa97/7
>>> 20p/mp4/file.mp4
>>>
>>> I shall write them and ask where they get those positive pulses.
>>
>> A transformer could invert pulses and take out the HV and the HV
>> ripple. You'd need to take out ps ripple for pulse height analysis.
>>
>
>Never thought of doing it like this but sounds reasonable.
>OTOH for PHA with a scintillator there is not much energy resolution
>to speak of so just about any shaping method you come up with will do.

One could also hang an opamp on the anode to get some gain before AC
coupling down to ground. A clever circuit would subtract out the power
supply ripple up there, or couple into a grounded diffamp. There are
some cheap, low-capacitance dc/dc converters to power the opamp.

On Fri, 22 Jul 2022 09:15:55 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

>Mike Monett wrote:
>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>
>> [...]
>>>> I shall write them and ask where they get those positive pulses.
>>>
>>> Probably from a TIA.
>>>
>>> One approach is to use an old-fashioned low-gain MMIC--most are
>>> basically Darlingtons with resistive feedback, so they invert.
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>
>> I begin to suspect a simple inverter. This shows the schematic of the
>> Theremino about 1/3 of the way down. 100 Meg feedback to a flyback HV
>> generator, 2 Meg PMT load, grounded cathode, pulse-shaping network to a
>> grounded emitter inverter. No concern about linearity. One thing I don't
>> understand is a 1N4148 connected betwen the base and ground to catch
>> negative pulses. I wonder if this is a cheap log converter:
>>
>> https://physicsopenlab.org/2016/01/26/diy-gamma-spectrometry/
>>
>> I think I'll run mine with a negative cathode to avoid mistakes wiping out
>> the preamplifier. It's not going to be in operation long enough to worry
>> about ion migration. Then find out how RH inverts their pulses.
>>
>>
>
>Looks like it's for short circuit protection. A spark across the output
>terminals could be quite unpleasant for T3 otherwise.
>
>Cheers
>
>Phil Hobbs

T3 looks well protected to me. The problem is that it should be Q3. T
means transformer.

On 22.7.22 15.56, Mike Monett wrote:
> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
> [...]
>>> I shall write them and ask where they get those positive pulses.
>>
>> Probably from a TIA.
>>
>> One approach is to use an old-fashioned low-gain MMIC--most are
>> basically Darlingtons with resistive feedback, so they invert.
>>
>> Cheers
>>
>> Phil Hobbs
>
> I begin to suspect a simple inverter. This shows the schematic of the
> Theremino about 1/3 of the way down. 100 Meg feedback to a flyback HV
> generator, 2 Meg PMT load, grounded cathode, pulse-shaping network to a
> grounded emitter inverter. No concern about linearity. One thing I don't
> understand is a 1N4148 connected betwen the base and ground to catch
> negative pulses. I wonder if this is a cheap log converter:
>
> https://physicsopenlab.org/2016/01/26/diy-gamma-spectrometry/
>
> I think I'll run mine with a negative cathode to avoid mistakes wiping out
> the preamplifier. It's not going to be in operation long enough to worry
> about ion migration. Then find out how RH inverts their pulses.

The diode is there to prevent the base-emitter diode from zenering,
and also to protect from large negative pulses.

--

-TV

On 7/22/2022 17:11, jlarkin@highlandsniptechnology.com wrote:
> On Fri, 22 Jul 2022 12:30:37 +0300, Dimiter_Popoff <dp@tgi-sci.com>
> wrote:
>
>> On 7/22/2022 3:17, John Larkin wrote:
>>> On Thu, 21 Jul 2022 23:53:53 -0000 (UTC), Mike Monett <spamme@not.com>
>>> wrote:
>>>
>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>
>>>>> Mike Monett wrote:
>>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>>
>>>>>> [...]
>>>>>>
>>>>>>> Yup. Back in the long ago (1984ish), I built a power supply filter
>>>>>>> that got rid of 10 Vpp of 120 Hz ripple on a 2 kV supply for some
>>>>>>> piezo stacks. It worked by lifting the cold end, sensing the hot end
>>>>>>> via a 100 nF, 3 kV film cap, and wiggling the cold end to keep it
>>>>>>> still. (It had some TVS protection too, obviously.)
>>>>>>>
>>>>>>> With an LF356 op amp, I got 100 dB of ripple rejection, which came in
>>>>>>> pretty handy. I could probably have done considerably better with
>>>>>>> that feedforward trick of Woodward's, where you use another op amp to
>>>>>>> measure the error voltage of the main one, and add in its output.
>>>>>>>
>>>>>>> Another approach like that is the Kanner Kap, which wasn't first
>>>>>>> invented by Kanner, but he now owns it on account of the cute name.
>>>>>>> That's the trick where you use an RC lowpass, and drive the cold end
>>>>>>> of the cap with some super beefy amplifier to make the top stay still.
>>>>>>>
>>>>>>> Cheers
>>>>>>>
>>>>>>> Phil Hobbs
>>>>>>
>>>>>> The CCFL inverter I got from Amazon runs at 40KHz.
>>>>>>
>>>>>> A simple half-wave rectifier running at 40 KHz with a 1nF cap and 9e6
>>>>>> Ohm load will produce about 2V sawtooth ripple.
>>>>>>
>>>>>> This is about 2/800 = 0.0025 * 100 = 0.25% ripple.
>>>>>>
>>>>>> That should be good enough.
>>>>>
>>>>> Totally. You can probably use one of the HV caps that came with it to
>>>>> couple the anode pulses to the outside world.
>>>>>
>>>>> Cheers
>>>>>
>>>>> Phil Hobbs
>>>>
>>>> That is a problem. The RH Electronics MCA expects positive pulses. These
>>>> won't come from the anode.
>>>>
>>>> https://static.wixstatic.com/media/e43988_a63520f4ed07436a843ddc1be0fda46a~
>>>> mv2.jpg
>>>>
>>>> https://video.wixstatic.com/video/e43988_8006efb319884c72b9d7f7418abdfa97/7
>>>> 20p/mp4/file.mp4
>>>>
>>>> I shall write them and ask where they get those positive pulses.
>>>
>>> A transformer could invert pulses and take out the HV and the HV
>>> ripple. You'd need to take out ps ripple for pulse height analysis.
>>>
>>
>> Never thought of doing it like this but sounds reasonable.
>> OTOH for PHA with a scintillator there is not much energy resolution
>> to speak of so just about any shaping method you come up with will do.
>
> One could also hang an opamp on the anode to get some gain before AC
> coupling down to ground. A clever circuit would subtract out the power
> supply ripple up there, or couple into a grounded diffamp. There are
> some cheap, low-capacitance dc/dc converters to power the opamp.
>

No need to do it that complicated, there just is no energy resolution to
worry about too much. They measure it in %, FWHM at the 661.6 137Cs peak
divided by 661.6. NaI detectors can go down to around 6%, usually 8 or
even 10 is still acceptable. This is 40keV FWHM at best, out of a
spectrum well within 2 MeV, more like 1.5. People often measure
1k spectra with NaI where clearly 128 channels would be plenty.
(Compare that to something like 1.3keV FWHM for the 137Cs line
with a Ge detector, with these you need 8 or 16k to get the full
picture).

I did a PMT circuit for photon counting a few years ago for a TLD
reader I designed, http://tgi-sci.com/tgi/tld/index.htm (no PHA).
Since the PMT glass hangs in the air there was no issue biasing it
negatively and coupling the anode into one of the modern day fast
opamps ADI make, works pretty well - throughput being around 22
million photons/second.

But my first MCA card for NaI - designed >30 years ago, the so called
TISA card - was expecting only detectors with negative bias. I had no
idea what detectors could be etc., I just designed the MCA card - and
it had to deliver up to 3 mA as the detectors it must have been meant
for had a preamp bleeding the HV for power.... (it could do up
to 1200V/3mA, programmable).

More recently I had a customer who wanted to connect an NaI
detector to one of our netMCA-s. The detector had just one BNC,
for both +HV in and signal out. So I had to build a tiny splitter
box for it... Clearly one gets one more pole in such a configuration
but it is too slow and is practically immaterial. Worked with the
standard netMCA HV configuration (up to 5kV for Ge), though the
HV had to be set to something like 2kV to compensate for the drop
over around 30 megaohm in the output filter (Ge detectors consume
practically no current).

Mike Monett <spamme@not.com> wrote:

[..]

> I think I'll run mine with a negative cathode to avoid mistakes wiping
> out the preamplifier. It's not going to be in operation long enough to
> worry about ion migration. Then find out how RH inverts their pulses.

Got a reply from RH. They don't. You do:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Hello Mike,

To flip the pulse polarity you can use a classic inverting OP-AMP circuit.

The chip op-amp you will use has to be able to process fast PMT pulses.
Usually 10MHz chip is enough in most cases with NaI(Tl) probe.

Please note, the MCA input specifications require for input signal
amplitude of 0mV-3300mV

Thank you
Alex
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Now I have to figure out the input impedance of the inverter. I'll have to
fire up the PMT and see what amplitude the pulses are.

Fortunately, my Radium order just arrived so I will have some signals.

Unfortunately the decay steps are alpha and beta. The scintillator is
gamma, so I don't think Ra will help, but it drives my Radiacode crazy.
First time I have heard the alarm. No wonder it made the girls who painted
it on their bodies glow in the dark.

I have some volcanic rock on order. I'll have to bug John and Dave and find
out where they got their sources. This stuff is tough to find. They think
you want to make a bomb or something.

--
MRM

jlarkin@highlandsniptechnology.com wrote:
> On Fri, 22 Jul 2022 12:30:37 +0300, Dimiter_Popoff <dp@tgi-sci.com>
> wrote:
>
>> On 7/22/2022 3:17, John Larkin wrote:
>>> On Thu, 21 Jul 2022 23:53:53 -0000 (UTC), Mike Monett <spamme@not.com>
>>> wrote:
>>>
>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>
>>>>> Mike Monett wrote:
>>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>>
>>>>>> [...]
>>>>>>
>>>>>>> Yup. Back in the long ago (1984ish), I built a power supply filter
>>>>>>> that got rid of 10 Vpp of 120 Hz ripple on a 2 kV supply for some
>>>>>>> piezo stacks. It worked by lifting the cold end, sensing the hot end
>>>>>>> via a 100 nF, 3 kV film cap, and wiggling the cold end to keep it
>>>>>>> still. (It had some TVS protection too, obviously.)
>>>>>>>
>>>>>>> With an LF356 op amp, I got 100 dB of ripple rejection, which came in
>>>>>>> pretty handy. I could probably have done considerably better with
>>>>>>> that feedforward trick of Woodward's, where you use another op amp to
>>>>>>> measure the error voltage of the main one, and add in its output.
>>>>>>>
>>>>>>> Another approach like that is the Kanner Kap, which wasn't first
>>>>>>> invented by Kanner, but he now owns it on account of the cute name.
>>>>>>> That's the trick where you use an RC lowpass, and drive the cold end
>>>>>>> of the cap with some super beefy amplifier to make the top stay still.
>>>>>>>
>>>>>>> Cheers
>>>>>>>
>>>>>>> Phil Hobbs
>>>>>>
>>>>>> The CCFL inverter I got from Amazon runs at 40KHz.
>>>>>>
>>>>>> A simple half-wave rectifier running at 40 KHz with a 1nF cap and 9e6
>>>>>> Ohm load will produce about 2V sawtooth ripple.
>>>>>>
>>>>>> This is about 2/800 = 0.0025 * 100 = 0.25% ripple.
>>>>>>
>>>>>> That should be good enough.
>>>>>
>>>>> Totally. You can probably use one of the HV caps that came with it to
>>>>> couple the anode pulses to the outside world.
>>>>>
>>>>> Cheers
>>>>>
>>>>> Phil Hobbs
>>>>
>>>> That is a problem. The RH Electronics MCA expects positive pulses. These
>>>> won't come from the anode.
>>>>
>>>> https://static.wixstatic.com/media/e43988_a63520f4ed07436a843ddc1be0fda46a~
>>>> mv2.jpg
>>>>
>>>> https://video.wixstatic.com/video/e43988_8006efb319884c72b9d7f7418abdfa97/7
>>>> 20p/mp4/file.mp4
>>>>
>>>> I shall write them and ask where they get those positive pulses.
>>>
>>> A transformer could invert pulses and take out the HV and the HV
>>> ripple. You'd need to take out ps ripple for pulse height analysis.
>>>
>>
>> Never thought of doing it like this but sounds reasonable.
>> OTOH for PHA with a scintillator there is not much energy resolution
>> to speak of so just about any shaping method you come up with will do.
>
> One could also hang an opamp on the anode to get some gain before AC
> coupling down to ground. A clever circuit would subtract out the power
> supply ripple up there, or couple into a grounded diffamp. There are
> some cheap, low-capacitance dc/dc converters to power the opamp.
> n

Well, the PMT has a gain of a million already. ;)

BTW the gain of a PMT goes as some high power (like 8 or 11 or
something) of the bias voltage.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

Mike Monett wrote:
> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
> [...]
>
>> Looks like it's for short circuit protection. A spark across the output
>> terminals could be quite unpleasant for T3 otherwise.
>>
>> Cheers
>>
>> Phil Hobbs
>
> That is precisely why I do not want to run mine with 800 Volts on the anode.
>
>
>
Why? It's super easy to protect against. There's no energy involved to
speak of.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

On Fri, 22 Jul 2022 16:48:16 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

>jlarkin@highlandsniptechnology.com wrote:
>> On Fri, 22 Jul 2022 12:30:37 +0300, Dimiter_Popoff <dp@tgi-sci.com>
>> wrote:
>>
>>> On 7/22/2022 3:17, John Larkin wrote:
>>>> On Thu, 21 Jul 2022 23:53:53 -0000 (UTC), Mike Monett <spamme@not.com>
>>>> wrote:
>>>>
>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>
>>>>>> Mike Monett wrote:
>>>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>>>
>>>>>>> [...]
>>>>>>>
>>>>>>>> Yup. Back in the long ago (1984ish), I built a power supply filter
>>>>>>>> that got rid of 10 Vpp of 120 Hz ripple on a 2 kV supply for some
>>>>>>>> piezo stacks. It worked by lifting the cold end, sensing the hot end
>>>>>>>> via a 100 nF, 3 kV film cap, and wiggling the cold end to keep it
>>>>>>>> still. (It had some TVS protection too, obviously.)
>>>>>>>>
>>>>>>>> With an LF356 op amp, I got 100 dB of ripple rejection, which came in
>>>>>>>> pretty handy. I could probably have done considerably better with
>>>>>>>> that feedforward trick of Woodward's, where you use another op amp to
>>>>>>>> measure the error voltage of the main one, and add in its output.
>>>>>>>>
>>>>>>>> Another approach like that is the Kanner Kap, which wasn't first
>>>>>>>> invented by Kanner, but he now owns it on account of the cute name.
>>>>>>>> That's the trick where you use an RC lowpass, and drive the cold end
>>>>>>>> of the cap with some super beefy amplifier to make the top stay still.
>>>>>>>>
>>>>>>>> Cheers
>>>>>>>>
>>>>>>>> Phil Hobbs
>>>>>>>
>>>>>>> The CCFL inverter I got from Amazon runs at 40KHz.
>>>>>>>
>>>>>>> A simple half-wave rectifier running at 40 KHz with a 1nF cap and 9e6
>>>>>>> Ohm load will produce about 2V sawtooth ripple.
>>>>>>>
>>>>>>> This is about 2/800 = 0.0025 * 100 = 0.25% ripple.
>>>>>>>
>>>>>>> That should be good enough.
>>>>>>
>>>>>> Totally. You can probably use one of the HV caps that came with it to
>>>>>> couple the anode pulses to the outside world.
>>>>>>
>>>>>> Cheers
>>>>>>
>>>>>> Phil Hobbs
>>>>>
>>>>> That is a problem. The RH Electronics MCA expects positive pulses. These
>>>>> won't come from the anode.
>>>>>
>>>>> https://static.wixstatic.com/media/e43988_a63520f4ed07436a843ddc1be0fda46a~
>>>>> mv2.jpg
>>>>>
>>>>> https://video.wixstatic.com/video/e43988_8006efb319884c72b9d7f7418abdfa97/7
>>>>> 20p/mp4/file.mp4
>>>>>
>>>>> I shall write them and ask where they get those positive pulses.
>>>>
>>>> A transformer could invert pulses and take out the HV and the HV
>>>> ripple. You'd need to take out ps ripple for pulse height analysis.
>>>>
>>>
>>> Never thought of doing it like this but sounds reasonable.
>>> OTOH for PHA with a scintillator there is not much energy resolution
>>> to speak of so just about any shaping method you come up with will do.
>>
>> One could also hang an opamp on the anode to get some gain before AC
>> coupling down to ground. A clever circuit would subtract out the power
>> supply ripple up there, or couple into a grounded diffamp. There are
>> some cheap, low-capacitance dc/dc converters to power the opamp.
>> n
>
>Well, the PMT has a gain of a million already. ;)

Don't the dynodes wear out after some numbers of coulombs?

>
>BTW the gain of a PMT goes as some high power (like 8 or 11 or
>something) of the bias voltage.

I made it to the Louisiana State Science Fair with an alpha
scintillator, a 931A in a pipe with a bit of glow paint scraped from a
clock. First Place in Physics! The competition was mediocre.

The alphas made giant pulses, way above background.

Next year I made it to the Nationals in Baltimore with a CRT thing.
Didn't win anything but hung out with Amory Lovins.

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

> Mike Monett wrote:
>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>
>> [...]
>>
>>> Looks like it's for short circuit protection. A spark across the
>>> output terminals could be quite unpleasant for T3 otherwise.
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>
>> That is precisely why I do not want to run mine with 800 Volts on the
>> anode.
>>
>>
>>
> Why? It's super easy to protect against. There's no energy involved to
> speak of.
>
> Cheers
>
> Phil Hobbs

Personal preference. I don't like the idea of feeding 800 volts into
electronics. Caps break down, pc traces arc from moisture or dirt, things
spark over as you mentioned earlier.

I'll run my cathode at -800 volts where nothing can be damaged.

Incidentally, a 10nF cap discharges from 800V to zero in about 50ms.

People say modern leds are crazy brilliant. I wonder if one will glow at
90uA to make a safety warning light that high voltage is present.

People often complain about leds staying lit on leakage currents in wiring,
but nobody measures the current.

--
MRM

John Larkin wrote:
> On Fri, 22 Jul 2022 16:48:16 -0400, Phil Hobbs
> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>> jlarkin@highlandsniptechnology.com wrote:
>>> On Fri, 22 Jul 2022 12:30:37 +0300, Dimiter_Popoff <dp@tgi-sci.com>
>>> wrote:
>>>
>>>> On 7/22/2022 3:17, John Larkin wrote:
>>>>> On Thu, 21 Jul 2022 23:53:53 -0000 (UTC), Mike Monett <spamme@not.com>
>>>>> wrote:
>>>>>
>>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>>
>>>>>>> Mike Monett wrote:
>>>>>>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>>>>
>>>>>>>> [...]
>>>>>>>>
>>>>>>>>> Yup. Back in the long ago (1984ish), I built a power supply filter
>>>>>>>>> that got rid of 10 Vpp of 120 Hz ripple on a 2 kV supply for some
>>>>>>>>> piezo stacks. It worked by lifting the cold end, sensing the hot end
>>>>>>>>> via a 100 nF, 3 kV film cap, and wiggling the cold end to keep it
>>>>>>>>> still. (It had some TVS protection too, obviously.)
>>>>>>>>>
>>>>>>>>> With an LF356 op amp, I got 100 dB of ripple rejection, which came in
>>>>>>>>> pretty handy. I could probably have done considerably better with
>>>>>>>>> that feedforward trick of Woodward's, where you use another op amp to
>>>>>>>>> measure the error voltage of the main one, and add in its output.
>>>>>>>>>
>>>>>>>>> Another approach like that is the Kanner Kap, which wasn't first
>>>>>>>>> invented by Kanner, but he now owns it on account of the cute name.
>>>>>>>>> That's the trick where you use an RC lowpass, and drive the cold end
>>>>>>>>> of the cap with some super beefy amplifier to make the top stay still.
>>>>>>>>>
>>>>>>>>> Cheers
>>>>>>>>>
>>>>>>>>> Phil Hobbs
>>>>>>>>
>>>>>>>> The CCFL inverter I got from Amazon runs at 40KHz.
>>>>>>>>
>>>>>>>> A simple half-wave rectifier running at 40 KHz with a 1nF cap and 9e6
>>>>>>>> Ohm load will produce about 2V sawtooth ripple.
>>>>>>>>
>>>>>>>> This is about 2/800 = 0.0025 * 100 = 0.25% ripple.
>>>>>>>>
>>>>>>>> That should be good enough.
>>>>>>>
>>>>>>> Totally. You can probably use one of the HV caps that came with it to
>>>>>>> couple the anode pulses to the outside world.
>>>>>>>
>>>>>>> Cheers
>>>>>>>
>>>>>>> Phil Hobbs
>>>>>>
>>>>>> That is a problem. The RH Electronics MCA expects positive pulses. These
>>>>>> won't come from the anode.
>>>>>>
>>>>>> https://static.wixstatic.com/media/e43988_a63520f4ed07436a843ddc1be0fda46a~
>>>>>> mv2.jpg
>>>>>>
>>>>>> https://video.wixstatic.com/video/e43988_8006efb319884c72b9d7f7418abdfa97/7
>>>>>> 20p/mp4/file.mp4
>>>>>>
>>>>>> I shall write them and ask where they get those positive pulses.
>>>>>
>>>>> A transformer could invert pulses and take out the HV and the HV
>>>>> ripple. You'd need to take out ps ripple for pulse height analysis.
>>>>>
>>>>
>>>> Never thought of doing it like this but sounds reasonable.
>>>> OTOH for PHA with a scintillator there is not much energy resolution
>>>> to speak of so just about any shaping method you come up with will do.
>>>
>>> One could also hang an opamp on the anode to get some gain before AC
>>> coupling down to ground. A clever circuit would subtract out the power
>>> supply ripple up there, or couple into a grounded diffamp. There are
>>> some cheap, low-capacitance dc/dc converters to power the opamp.
>>> n
>>
>> Well, the PMT has a gain of a million already. ;)
>
> Don't the dynodes wear out after some numbers of coulombs?

About 1000 coulombs per square centimeter of photocathode, give or take.
They also age out after five years or so because all the volatile
stuff like caesium metal migrates around inside. If a gamma counter
gets up to 1000 C/cm**2, I suspect Mike would have other problems. ;)

>
>>
>> BTW the gain of a PMT goes as some high power (like 8 or 11 or
>> something) of the bias voltage.
>
> I made it to the Louisiana State Science Fair with an alpha
> scintillator, a 931A in a pipe with a bit of glow paint scraped from a
> clock. First Place in Physics! The competition was mediocre.

Fun.

>
> The alphas made giant pulses, way above background.

Yup.

>
> Next year I made it to the Nationals in Baltimore with a CRT thing.
> Didn't win anything but hung out with Amory Lovins.

Don't know him.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

Mike Monett <spamme@not.com> wrote:

> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>> Why? It's super easy to protect against. There's no energy involved to
>> speak of.
>>
>> Cheers
>>
>> Phil Hobbs
>
> Personal preference. I don't like the idea of feeding 800 volts into
> electronics. Caps break down, pc traces arc from moisture or dirt,
> things spark over as you mentioned earlier.

It doesn't matter which end gets the high voltage. Why take chances by
putting the high voltage on the anode. It works just as well by putting the
voltage on the cathode. I'm not going to be using it long enough to worry
about ion migration.

Toyota Principle. Poka-Yoke: Catch mistakes before they occur.

--
MRM

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

> jlarkin@highlandsniptechnology.com wrote:

[...]

>> One could also hang an opamp on the anode to get some gain before AC
>> coupling down to ground. A clever circuit would subtract out the power
>> supply ripple up there, or couple into a grounded diffamp. There are
>> some cheap, low-capacitance dc/dc converters to power the opamp.
>
> Well, the PMT has a gain of a million already. ;)
>
> BTW the gain of a PMT goes as some high power (like 8 or 11 or
> something) of the bias voltage.
>
> Cheers
>
> Phil Hobbs

The high sensitivity means the HV supply has to be well regulated.

One site mentioned the HV has to be adjusted for each PMT to get the
spectrum lines to align with the corresponding MeV values, which makes
sense. I suppose the system would have to be recalibrated as the PMT ages.

--
MRM

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

[...]
> BTW the gain of a PMT goes as some high power (like 8 or 11 or
> something) of the bias voltage.
>
> Cheers
>
> Phil Hobbs

I wonder if that is because of the mechanism of operation, and the large
number of cascaded sections.

As I understand it, electrons hit a dynode and eject a number of electrons.
These electrons accelerated under the influence of the voltage between
electrodes, the same as in the old vacuum tubes. They hit the next
electrode and eject more electrons. The number of electrons depends on the
impact velocity, so the higher the voltage, the more electrons you get.

This process is repeated for each dynode, so the effect is multiplied each
time. This is why the PMT is so sensitive to the applied voltage.

And this is why it is so difficult to keep the spectrum aligned with the
known energy levels.

--
MRM

On 22/07/2022 14:07, Phil Hobbs wrote:
> Martin Brown wrote:
>>
>> Life improved considerably when they moved onto turbo-molecular pumps
>> and for a while performance was improved even more by maglev ones.
>>
>> Then one day an earthquake in Tokyo suddenly moved the Earth sideways
>> by an inch instantaneously and every damn one of them was destroyed. I
>> remember it well since in the quiet of the evening I heard the
>> earthquake coming (I have no idea how but I was wondering about the
>> strange train like sound outside when the jolt suddenly arrived).
>>
>> We were busy doing turbo pump swaps for weeks and after that
>> conventional bearings were specified for earthquake zones.
>>
>> The maglevs had been out in the field for a couple of years before
>> this rather nasty vulnerability became apparent.
>>
>
> Did any of the casings let go?

No. By then they had failure modes well under control.

The only time I have ever known it happen was with a much bigger high
power early model turbo pump (one of the largest available at the time).
Fortunately it failed at lunchtime and there was no-one in the lab. It
was pumping a big chamber about 1m^3 and something went horribly wrong.
The academic experimental setup was intrinsically not fail safe.

It shredded the blackout curtains, imploded the lab windows and doors.
The bang was heard all over the site and people went running to see what
had happened. Titanium rotor blades were stuck about 2" into concrete,
and went through breezeblock walls in a worryingly large kill zone. Some
were also stuck in the casing walls. Miraculously no-one was harmed.

The replacement included external Kevlar armour and a host of other
safety measures to prevent a recurrence from harming experimenters.

--
Regards,
Martin Brown

Mike Monett wrote:
> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
> [...]
>
>> BTW the gain of a PMT goes as some high power (like 8 or 11 or
>> something) of the bias voltage.
>>
>> Cheers
>>
>> Phil Hobbs
>
> I wonder if that is because of the mechanism of operation, and the
> large number of cascaded sections.

>
> As I understand it, electrons hit a dynode and eject a number of
> electrons. These electrons accelerated under the influence of the
> voltage between electrodes, the same as in the old vacuum tubes. They
> hit the next electrode and eject more electrons. The number of
> electrons depends on the impact velocity, so the higher the voltage,
> the more electrons you get.
>
> This process is repeated for each dynode, so the effect is
> multiplied each time. This is why the PMT is so sensitive to the
> applied voltage.

Yes, that's right. The secondary yield generally isn't quite
proportional to the bias per stage, because at higher bias the incident
electrons penetrate deeper, so that their secondaries have to make their
way out through a thicker layer of material on average, and still have
enough energy to escape when they hit the surface.

There are negative electron affinity (NEA) photocathodes, where
essentially any free electron that reaches the surface will escape.
They have higher quantum yield, as you'd expect, but are also slower
because it takes awhile for all those low-energy secondaries to bounce
around before reaching the surface. I'd expect tubes with NEA dynodes
to follow the power law more closely.

>
> And this is why it is so difficult to keep the spectrum aligned with
> the known energy levels.

Well, and the fact that the energy resolution of CsI is the pits
compared with something like intrinsic germanium. On the plus side,
it's cheap and doesn't need liquid nitrogen.

Back in the days before really pure silicon and germanium crystals were
available, they used to be "lithium drifted", i.e. lithium interstitials
were introduced to bind to impurity trap states. Those ones had to be
kept at 77K continuously, or else they'd be spoiled in a few days.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

i

Mike Monett wrote:
> Mike Monett <spamme@not.com> wrote:
>
>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>>> Why? It's super easy to protect against. There's no energy involved to
>>> speak of.
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>
>> Personal preference. I don't like the idea of feeding 800 volts into
>> electronics. Caps break down, pc traces arc from moisture or dirt,
>> things spark over as you mentioned earlier.
>
> It doesn't matter which end gets the high voltage. Why take chances by
> putting the high voltage on the anode. It works just as well by putting the
> voltage on the cathode. I'm not going to be using it long enough to worry
> about ion migration.
>
> Toyota Principle. Poka-Yoke: Catch mistakes before they occur.
>
>
>
Provided you can insulate the can from the photocathode.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

On Sat, 23 Jul 2022 08:14:52 -0000 (UTC), Mike Monett <spamme@not.com>
wrote:

>Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>[...]
>
>> BTW the gain of a PMT goes as some high power (like 8 or 11 or
>> something) of the bias voltage.
>>
>> Cheers
>>
>> Phil Hobbs
>
>I wonder if that is because of the mechanism of operation, and the large
>number of cascaded sections.
>
>As I understand it, electrons hit a dynode and eject a number of electrons.
>These electrons accelerated under the influence of the voltage between
>electrodes, the same as in the old vacuum tubes. They hit the next
>electrode and eject more electrons. The number of electrons depends on the
>impact velocity, so the higher the voltage, the more electrons you get.
>
>This process is repeated for each dynode, so the effect is multiplied each
>time. This is why the PMT is so sensitive to the applied voltage.
>
>And this is why it is so difficult to keep the spectrum aligned with the
>known energy levels.

Add some source with a known narrow line, and servo on that.

Martin Brown wrote:
> On 22/07/2022 14:07, Phil Hobbs wrote:
>> Martin Brown wrote:
>>>
>>> Life improved considerably when they moved onto turbo-molecular pumps
>>> and for a while performance was improved even more by maglev ones.
>>>
>>> Then one day an earthquake in Tokyo suddenly moved the Earth sideways
>>> by an inch instantaneously and every damn one of them was destroyed.
>>> I remember it well since in the quiet of the evening I heard the
>>> earthquake coming (I have no idea how but I was wondering about the
>>> strange train like sound outside when the jolt suddenly arrived).
>>>
>>> We were busy doing turbo pump swaps for weeks and after that
>>> conventional bearings were specified for earthquake zones.
>>>
>>> The maglevs had been out in the field for a couple of years before
>>> this rather nasty vulnerability became apparent.
>>>
>>
>> Did any of the casings let go?
>
> No. By then they had failure modes well under control.
>
> The only time I have ever known it happen was with a much bigger high
> power early model turbo pump (one of the largest available at the time).
> Fortunately it failed at lunchtime and there was no-one in the lab. It
> was pumping a big chamber about 1m^3 and something went horribly wrong.
> The academic experimental setup was intrinsically not fail safe.
>
> It shredded the blackout curtains, imploded the lab windows and doors.
> The bang was heard all over the site and people went running to see what
> had happened. Titanium rotor blades were stuck about 2" into concrete,
> and went through breezeblock walls in a worryingly large kill zone. Some
> were also stuck in the casing walls. Miraculously no-one was harmed.
>
> The replacement included external Kevlar armour and a host of other
> safety measures to prevent a recurrence from harming experimenters.
>
Yikes.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

On 7/23/2022 11:30, Martin Brown wrote:
> On 22/07/2022 14:07, Phil Hobbs wrote:
>> Martin Brown wrote:
>>>
>>> Life improved considerably when they moved onto turbo-molecular pumps
>>> and for a while performance was improved even more by maglev ones.
>>>
>>> Then one day an earthquake in Tokyo suddenly moved the Earth sideways
>>> by an inch instantaneously and every damn one of them was destroyed.
>>> I remember it well since in the quiet of the evening I heard the
>>> earthquake coming (I have no idea how but I was wondering about the
>>> strange train like sound outside when the jolt suddenly arrived).
>>>
>>> We were busy doing turbo pump swaps for weeks and after that
>>> conventional bearings were specified for earthquake zones.
>>>
>>> The maglevs had been out in the field for a couple of years before
>>> this rather nasty vulnerability became apparent.
>>>
>>
>> Did any of the casings let go?
>
> No. By then they had failure modes well under control.
>
> The only time I have ever known it happen was with a much bigger high
> power early model turbo pump (one of the largest available at the time).
> Fortunately it failed at lunchtime and there was no-one in the lab. It
> was pumping a big chamber about 1m^3 and something went horribly wrong.
> The academic experimental setup was intrinsically not fail safe.
>
> It shredded the blackout curtains, imploded the lab windows and doors.
> The bang was heard all over the site and people went running to see what
> had happened. Titanium rotor blades were stuck about 2" into concrete,
> and went through breezeblock walls in a worryingly large kill zone. Some
> were also stuck in the casing walls. Miraculously no-one was harmed.
>
> The replacement included external Kevlar armour and a host of other
> safety measures to prevent a recurrence from harming experimenters.
>

Whoa! Having no people in the room makes that a fun story, the imploded
windows, shredded curtains etc. - hard to match. My closest electronics
related event compares to that like stumbling in the street to a plane
crash, I am somewhat envious :).
[Once, almost 30 years ago, I reversed the power on a new board with
plenty of tantalum caps which exploded machine-gun like, pretty loud at
that. Lucy was in the next room and had frozen thinking she might have
lost me; unfroze with a huge sigh of relief some 10 seconds later when I
grasped what had happened and started to chain-swear].

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

[...]
> Cheers
>
> Phil Hobbs
Thanks very much for the lucid explanation.

--
MRM

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

> Mike Monett wrote:
>> Mike Monett <spamme@not.com> wrote:
>>
>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>
>>>> Why? It's super easy to protect against. There's no energy involved
>>>> to speak of.
>>>>
>>>> Cheers
>>>>
>>>> Phil Hobbs
>>>
>>> Personal preference. I don't like the idea of feeding 800 volts into
>>> electronics. Caps break down, pc traces arc from moisture or dirt,
>>> things spark over as you mentioned earlier.
>>
>> It doesn't matter which end gets the high voltage. Why take chances by
>> putting the high voltage on the anode. It works just as well by putting
>> the voltage on the cathode. I'm not going to be using it long enough to
>> worry about ion migration.
>>
>> Toyota Principle. Poka-Yoke: Catch mistakes before they occur.
>>
>>
>>
> Provided you can insulate the can from the photocathode.
>
> Cheers
>
> Phil Hobbs

I'll find out soon. Thanks

--
MRM

On Fri, 22 Jul 2022 23:39:20 -0000 (UTC), Mike Monett <spamme@not.com>
wrote:

>Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>> Mike Monett wrote:
>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>
>>> [...]
>>>
>>>> Looks like it's for short circuit protection. A spark across the
>>>> output terminals could be quite unpleasant for T3 otherwise.
>>>>
>>>> Cheers
>>>>
>>>> Phil Hobbs
>>>
>>> That is precisely why I do not want to run mine with 800 Volts on the
>>> anode.
>>>
>>>
>>>
>> Why? It's super easy to protect against. There's no energy involved to
>> speak of.
>>
>> Cheers
>>
>> Phil Hobbs
>
>Personal preference. I don't like the idea of feeding 800 volts into
>electronics. Caps break down, pc traces arc from moisture or dirt, things
>spark over as you mentioned earlier.
>
>I'll run my cathode at -800 volts where nothing can be damaged.
>
>Incidentally, a 10nF cap discharges from 800V to zero in about 50ms.
>
>People say modern leds are crazy brilliant. I wonder if one will glow at
>90uA to make a safety warning light that high voltage is present.
>
>People often complain about leds staying lit on leakage currents in wiring,
>but nobody measures the current.

90 uA should be pretty visible into a decent LED. Some are visible in
room light at 1 uA.

Here is my high-voltage blinker:

https://www.dropbox.com/s/53vtce00svlwur4/Blink_1.jpg?raw=1

It makes bright flashes.

On Sunday, June 26, 2022 at 6:53:17 PM UTC-4, John Miles, KE5FX wrote:
> On Saturday, June 25, 2022 at 11:16:45 AM UTC-7, Mike Monett wrote:
> > I tried to order one. Does not ship to Canada. Can you order one and mail it
> > to me? I can pay you via Paypal. Thanks.
> Maybe. I'll have to surf through the Commerce and ITAR lists to see if there's
> a reason why the seller doesn't ship to Canada. Shifty folk, Canadians.
>
> (It'll be a few days before I have time to deal with it, for various reasons.)
>
> -- john, KE5FX

There are a few reasons we do not ship internationally. Main one is ITAR and having to figure out what is and isn't can be a pain and it's not worth it for us. Second would be shipping issues. We have no control over the shipping and a lot of items would get lost in the past and then it's on us to refund. I know there are some new ebay programs which reduce this risk and it is something we are considering for the future. I am really happy we were able to save this tube from the scrap heap and that so many are finding it useful.