X-Ray Systems

Version 1.00

Copyright (C) 1999
Samuel M. Goldwasser
--- All Rights Reserved ---

For contact info, please see the Sci.Electronics.Repair FAQ Email Links Page.


Reproduction of this document in whole or in part is permitted if both of the following conditions are satisfied:
  1. This notice is included in its entirety at the beginning.
  2. There is no charge except to cover the costs of copying.


Table of Contents

X-Ray System Safety

(From: Terry Greene (xray@cstel.net).)

If an X-ray power supply is fired unloaded, output voltage can exceed a quarter of a million volts, and can hold an incredible charge long after the power is disconnected and can easily kill. I've startled people down the hall by shorting the generator secondaries on a disconnected machine that was fired (malfunction) unloaded. An impressive discharge. NEVER work on an x-ray supply without shorting the output first. If it doesn't kill you, you won't forget it... assuming you ever remember it. :-)

There is no practical problem with x-radiation at the typical voltage used in most gas type lasers. All mammography tubes use special beryllium windows for the radiation output because x-radiation at those low energies (20 to 32 kvp) won't penetrate the glass envelope of a standard tube. I've tried it. A glass envelope (standard) x-ray tube won't produce any exposure at all at 15 kvp according to my test equipment. (Keithly ion chamber) Even with complete evacuation, I seriously doubt you could find measurable x-radiation output from a glass bore laser.

Medical and Dental X-Ray Equipment

Basic Characteristics

The following deals with medical and dental X-ray machines. See the section: Typical Computed Tomography X-Ray Generators for information on that higher power equipment.

(From: Terry Greene (xray@cstel.net).)

Dental X-ray usually has fixed kVp in the 65 to 90 kVp range. The vast majority of them are set up at 70 kVp, usually at about 10 ma.

As far as small medical, hmmm... define small. General medical stuff is usually adjustable in the 50 to 125 kVp range for single phase or 50 to 150 kVp for three-phase. These machines usually have adjustable mA from 25 to as much as 1000 mA. Smaller units as found in the offices of general practitioners will typically be adjustable from 50 to 300 ma.

A few of the many exceptions to the above:

  1. Portables:

  2. Mammography:

  3. High frequency:

    Operating at either 20 kHz or 100 kHz, these unit have rms output that exceeds three phase output by a few percent.

Typical Failure Modes of X-Ray Tubes

Most common is a fried anode from putting more power to the tube than it will handle faster than it can disipate it. The track where the beam hits will stress fracture and crack up. It will look like an alligators back. When the happens, the mr per mas will drop and the film density will drop proportionally. Detail will also drop some what as the focal spot is now effectively not stationary as the beam walks up and down the irregularities. If the anode is seriously overheated when the anode is cold it can crack wide open from the edge to the shaft. I have seen one that broke clean in two. Next would probably be bad (noisy) bearings. I've seen them run noisy for years so those are perfect for a hobby machine. You hear a lot about "gassed" tubes, but rarely see them. Most tubes that are classified as "gassy" simply need to be reseasoned. Where are you? Shipping can be high on the large units.

Typical X-Ray Dose

(From: Jeff (jeffro5@roanoke.infi.net).)

I enjoy reading discussions that add to our knowledge but I immediately became suspicious of the dose values reported by the two previous gentlemen.

After some quick research to confirm the values I had in mind, here's what I found:

(From: Gregory W. Froehlich (Gregory.W.Froehlich@dartmouth.edu).)

In 1981, the mean exposure for bitewing dental films was 334 mR. It's certainly less now, so 70 mR doesn't seem too surprising (in 1980, a chest film was about 30 mR, and a mammogram about 500).

The stuff I've seen for total average annual radiation is about 300 mR. Background radiation (environment and from people's own bodies) accounts for about 80% of that. Radon, on average, makes up about 55% of the total. Now, if you live in parts of New England (high radon, from the granite I think), the background radiation is 2.5 times as high; if you live in Leadville Colorado (lots of cosmic rays and metal ores), the background level is 3 times as high. Don't know about the exposure per hour of sun exposure; it's probably less than that of sitting inside, since radon levels are higher indoors and the main problem with sun is UV, not ionizing radiation (X or gamma). Anybody know the exposure rate for flying at 33,000'? I'd bet a coast-to-coast trip comes with 100 mR or more, from cosmic rays.

X-rays don't spatter, but they do scatter. Some x-rays are partially absorbed by the patient's body and are scattered. This normally would cause image unsharpness, but most modern systems also employ a focused grid under the patient to filter out scattered rays and only allow direct x-rays to pass. Incidentially, these scattered x-rays are what cause the occupational exposures to x-ray personnel, however their intensity is usually less than 1% of the primary beam intensity at 1 meter from the central beam axis.

One last important factor in x-ray imaging science is the X-ray tube's focal spot dimensions. A smaller focal spot will give a sharper resultant image. Normal modern x-ray machines usually have two focal spots which range between 0.8mm and 2.0 mm. In mammography, where image sharpness is particularly important, standard focal spot sizes are 0.1 mm and 0.3 mm.

The next step, which is occurring now, will be the replacement of film entirely with digital imaging receptors. This will further lower patient dose, minimise retakes, and give added imaging quality as the technology advances into the future.

For more about X-ray imaging physics, I would recommend "Physics of Radiology" by Anthony B. Wolbarst, 1993, Simon & Schuster.

Line Frequency and High Frequency X-Ray Generators

(From: Terry Greene.)

One difference is in how the HV obtained from line voltage:

A 3 phase 12 pulse system takes the line voltage, puts it through step up transformer(s), rectifies it, and applies it to the X-ray tube. Very simple technology that dates to the early days of commercial X-ray systems.

A high frequency generator takes the line voltage rectifies and filters to to make DC. The DC is then chopped at a high frequency and a combination of transformers and voltage multipliers then steps it up to the required voltage. Using a high frequency enables many of the components to be much smaller such as any transformers and capacitors.

High frequency generators are smaller and lighter and lend themselves to digital control.

High frequency generators are not used eexclusively with digital X-ray system. It is simply an efficient, light weight means of producing the high voltage. For example, modern CT scanners almost always use this technology especially where the HV generator is mounted on the rotating gantry. (Unless you consider CT to be a subset of digital X-ray which it is). There is no technical reason why a high frequency generator cannot be used with any X-ray system.

(From: Chris Smolinski.)

I've designed a switching power supply which drives a HV multiplier inside of an X-ray source. The X-ray source contains a 1:26 step-up transformer and a multiplier to produce up to 80 kV. I need to feed it with a 12 kHz signal, about 300 V p-p. The design of the x-ray source cannot be changed, I'm stuck with it.

My power supply uses a push-pull amplifier to produce this 12 kHz signal. The center tap of the primary of a transformer is connected to a 96V DC supply, and I PWM the signal to two MOSFETS connected to each side of the primary. Maximum output power is about 500W.

I've noticed that after a period of operation at higher power levels, the transformer becomes quite warm [hot, I guess it's kinda relative]. I don't have exact temperatures handy. Perhaps I'm being too conservative, but I'd like the transformer to run somewhat cooler. The transformer design was mostly empirically determined, I'm sure it's far from optimum.

Suggestions on what I can do to improve it? Sources for useful design hints/information? Unfortunately, they don't seem to teach you anything about transformer design in EE courses anymore (at least not at U of Maryland 6 years ago...). From: STOICHITA Catalin Old X-ray machines and film (Was: Re: radiation dosages) Mike Bohan explained about the basics of X-rays imaging he tried to cover more than their the aplication in dentristy. Just few complementary remarks: 1) The X-rays generators for dentistry (intraoral pictures, panoramic pictures and teleradiography pictures) are only mono focal types. 2)A typical focal size for 5-10 years old pano's (and attached teleradio units) my be consideed as 1x1mm but the modern types are close to 0.6x0.6mm. For the intraoral imaging the focal size was from long time ago less than 0.7x0.7mm. 3) There is no usage of the focussed grid in dentistry. They are largely used in other fields of radiology. Brian Sandle asked about film sensitivity meaning. X-ray Film Sensitivity (From: STOICHITA Catalin (signet@club-internet.fr).)

In rare cases the film is used isolated (for exemple in retroalveolar examination). Generaly they are used toghether with fluorescent screens(" enhancing screens " or " phosphor screen ") The darkness (optical density) of the developed picture, is reflecting the sensitivty of overall system: film+fluorescent screen so, it is useless to speak separately about the " sesitivity of the new films ". One of the most known supplier is KODAK. They have a lot of little but very clear manuals about the radiology. Other important suppliers are AGFA, 3M, FUJI, DUPONT, SIEMENS.

The main parameters of the couple film+enhancing screen are sensitivity and resolution. The couples are specific to type of examinations because in some cases is very important to obtain maximum resolution and in other maximum sensitivity. More sensitive couple means less X-ray photons sent to the patient but do not forget the X-rays energy spectrum.

For essential the X-rays are giving a little part of the film darkness but the light from the screens is the major fillm darkenss source. This light is proportional to the X-ray absortion by the screen. The absortion is essentially dependent on the X-ray energy. If you choose two different couples and you compare their sensitivities in two sittuation, let's say at 80kV and at 15kV, you may find that each one is more sensitive than the other but at prefered kV.

Let's remember that the CaWO4 was very long time the main ( see only) phosphor used in radiology. Some products based on it were used as reference. This phosphor is practically obsolette today but unfortunatelly is still used as referance. Unfortunatelly there are two implications:

  1. When a dealer of films or screens is speaking about the speediness of his products he is forgetting to explain that obsolete reference or if he is writing, some times he note with very little fonts "100% speediness is the CaWO4 screen " with no other comment.

  2. The legal or recomanded X-rays dosage limits are establishied long time ago when it was not possible to do better and are left unchanged many times.
And, about using modern film on old X-rays machines, basicaly there is no restriction. However adiacent problems may arrive:
  1. Fitting actual films into a very old film holder model.
  2. Reducing the current below a given value.
  3. Reducing the time exposure on panoramics is not often easy.
  4. Is really interesting to use a VERY OLD X-rays machine?
Replacing the films by TV cameras is well known and under continue progress from long time ago. The thecnology that is relatively new is CCD based X-rays sensor. The first step was done by Trophy Radiologie (France) about 7 years ago when the RVG (RadioVisio Graphy) was introduced on the market. RVG is a CCD based X-ray sensor to be placed in the mouth rear the teeth just like the well known intraoral film. A cable links the sensor through a digital electronics to a display. The system is used like before, with the classical films but insted the chemical processing we have instant images on the display (computer). The first solution proposed for the intraoral sensor was for short time because 1992, Regam itroduced his X-rays intraoral imaging system named " Sens-a-Ray ". One year later, Gendex (Italy/USA) introduced the same technology but named " Visual X "One year late Shick (USA) introduced his sensors and then Siemens(Germany),and MedizinReichner (Germany). Soredex (Finland) proposed a solution based on an intermediary support: memory phosphor of Fuji.

A second application in radiology for the CCD's is for the digital panoramic. In 1995, SIGNET(France) begun the selling of DXIS . This is a kind of universal kit being able to upgrade any classical panoramic int a digital one. Siemens an Trophy followed with fixed configuration for their last pano model. Planmeca(Finland) is also announcing his digital panoramic. Some particularities merit to be underlined:

  1. Trophy sollution consists in replacement of the film holder of their (Instrumentarium-Finland made) OP100 panoramic with a " digital cassette " that is a kind of " film holder " which is in the fact an electronic device-the sensor. So, the usage of the film is still possible with this solution.

  2. Planmecca and Siemens propose a pure digital panoramic that is in the place of film subassembly there is a fixed sensor.

  3. The DXIS technology from SIGNET targets the global park of the existing pano's.
Let's return an instant to the sensitivity. When the intra oral sensors arrived, the commercials spoken: " The CCD based sensor is 4 times more sensitive as the film !" (But what film?! ) When the second arrived, other commercials said: " our CCD based sensor is 5 times more sensitive as the film " and so on... Other formulas which are equivalent as arithmetic but more penetrant were also used " our sensor permits to reduce by 75% the radiation! " or " our sensor permits to reduce by 80% the radiation " and so on... The rough commercial competition pused far the evaluation of intraoral sensors. Each manufacturer dicovered by pur hazard articles in the press which praise their product and immediatelly displayed it. Conclusion: a high amplitude wawe was created sustaining the thesys that the new CCD based technology permits to reduce many many times the Xrays.

As designer of one of these sensors and of the DXIS system, I am convinced that the main explanation for the CCD based X-rays intraoral sensor is not by the CCD usage but just by the usage of the phosphor screen. The commercials forgeted to explain that they are comparing the classical very high resolution dental film, which is not coupled with phosphor screens with a system based on CCD which is receiving the light from a phosphor screen. Near the same sensitivity may be obtained with a film coupled to the screens. In the fact they realised an important move on the market: they convinced the dentists that is better to renounce to the resolution which is too high and get benefit from the X-ray economy.

The most important advantage that the digital sensors bring is the computer environment.

The CCD story riscs to be repeted in the competition on panoramics. Subject: Re: How to Focus X-ray? (From: Tom Loredo (loredo@spacenet.tn.cornell.edu).) The ability of metals to reflect x-rays decreases greatly with the x-ray energy. I don't know the numbers off the top of my head (I last worked with a grazing incidence x-ray telescope over a decade ago), but no technology I'm aware of focuses gamma rays. Astronomical grazing incidence telescopes don't go much higher than 100 keV, I believe, if they even go that high. Imaging gamma ray telescopes currently simply occult most of the sky, and measure the actual angle of incidence of detected gamma rays via Compton scattering (e.g., the EGRET telescope on the Compton Gamma Ray Observatory). The system I worked on had two conic sections (parabola/parabola or parabola/hyperbola, I forget which) that both reflected from the *inside* surface to achieve focus. We nested three of them to build up the effective area. 2-d info about the position of the focused gamma ray in the focal plane was obtained using a microchannel plate and a 2-d resistor. By the way, this is not a do-it-at-home activity. Our telescope, for example, had gold-plated grazing incidence mirrors and the mirrors alone cost something like $40k. And since it's grazing incidence, the effective area is small. (From: Douglas Dwyer (ddwyer@ddwyer.demon.co.uk).) I thought I would hear all sorts of comments re this recently published technique. A team of researchers under Anatoly Snigirev at ESRF in fr have made use of the 2.8e-6 difference in refractive index between Al and air (air is higher) by creating a refractive focussing lens from a series of 2D lenses from cylindrical holes in Al. Each lens is in series and reduces the overall focal length. Seems simple how come no one thought of it before? :) What about creating a 3D lens by positioning air/nitrogen bubbles in a tapered array within a volume of Aluminum. > My latest dilemma concerns, as the title of this post would suggest, > the issue of potting for security and the x-raying of said pot for the > purpose of breaching that security. > My question may be out of the realm of electronic design, but since > there seem to be many *know it alls* (~: that frequent this newsgroup, > I thought I'd give it a shot here in SED. > Question is,,,,,, Would a copper clad circuit board, un-etched, and > coated with solder, be sufficient to block the x-rays from an x-ray > machine and foil the intentions of someone trying to hack the circuit by > means of x-raying ? (From: Bill sloman (bill.sloman@ieee.org).) Subject: Re: Potting and X-Rays ??????? It depends on what you are trying to conceal and from whom. X-ray absorbtion depends on atomic number. Carbon has an atomic number of 6, silicon 14, copper 29, tin 50, and lead 82, so the lead in the solder will dominate the picture. Wrapping the circuit in lead foil would probably be better, and loading the potting mix with lead shot of a variety of sizes would be even better. Any one of these would probably stop someone trying to get a shadow image of the circuit with a simple medical X-ray machine, and would probably cut down the transmission through the potted sample enough so that a brain scanner wouldn't get very far. Somebody with the resources to lash-up a high voltage X-ray source to a precision stage could probably improvise an effective tomographic set-up, and someone with access to neutron radiography might be able to see through the lead, but it would probably be easier to organise a break-in at your plant to steal your drawings. From: Terry Greene Nifty! One correction. In the 1st info I sent I told you that most battery powered portables were 20 khz high frequency machines. This is true of the newer machines. The older ones (G.E. AMX etc..) are generally 500 hz. Let me know what info you might need. I've got tech data on most any model around including schematics. Don't know if I've told you, but I feel that your FAQs are a major resource and I personally thank you for taking the time and effort to publish them. It's a great deal of selfless work. Later, Terry Sam Goldwasser wrote: > > Greetings: > > I've started to write the 'X-ray FAQ' which will include the info you have > sent in the past, may experience in disassembling the tube, and any other > X-ray tid-bits that I come across. I don't know when this will be in a form > to put up at my Web sites, however. > > --- sam From: Terry Greene Sam Goldwasser wrote: > There are 3 wires on the cathode end with continuity between them. > How many filaments and what voltage/current do they take? There are two filaments, one small focal spot and one large. One connection is the common. I'm not at the office so I don't have the data sheets for that tube in front of me for exact numbers, but the voltage varies around six volts @ a 3 to 5 amps. At 4 amps most tubes will run around 100 ma @ 50kv. All modern tubes have similar ratings. In the display I built I used a 6.3 V, 6 A transformer as that's what I had laying around. It worked just fine on the large filament. Might blow the small at that voltage though. The filaments are designed to control the current by temp (as you know) and generally run at a bright orange in operation. If it starts getting too close to white, it's too hot and you risk smoking it. Since you are building a display the voltage/current won't be critical, but I wouldn't let it go much over 5 amps to keep from toasting the filament. I know the Eureka you have has a max filament current listed of 5.2 amps. for each of the filaments with normal filament voltage (at max current) listed at 7.8 to 10.6 for the large, and 5.9 to 8.1 for the small. The GE won't be far from that. The motor in common 600 ma and lower systems spins at around 3,600 rpm. Many of the high power systems have a high frequency rotor drive circuits that will spin the rotor at 10,800 rpm. G.E. high speed rotor controllers are known as a RARC (Rapid Acceleration Rotor Controller) and SARC (Super Acceleration Rotor Controller). GE just loves acronyms. Any tube rotor will work fine on 60 hz. High speed 180 hz. operation just raises the power handling capability a bit. There are no standard color codes. A DVM will tell you all you need to know. Two windings. One connection is a common. One connection is a main winding at around 20 ohms. One connection is a phase shift at at around 50 ohms. If you check across the outside leads you will see around 70 ohms. Those are average numbers and the exact numbers may vary significantly from model to model, but the ratio will stay reasonably consistent. Some are as high as 30/90 with 120 total, some as low as 15/35 with 50 total. The common is hooked to line common, the main hooks to hot and the phase shift needs a motor capacitor in the line and hooks to hot as well. Around 30uf works well. The value is not critical. I've seen systems with 15uf. Medical X-ray systems usually start the rotor by applying 220VAC to spin the rotor up rapidly. After about 1.5 seconds the voltage is reduced to around 50VAC. Some of the inexpensive vet. units simply bring the rotor up on 120VAC and leave it at that. The acceleration is slightly slower, but it's barely noticeable. The vet. units that do so usually have a 2 sec delay before exposure is allowed instead of a 1.5 sec delay. All modern tube rotors will run at 120VAC all day without damage. For my display I simply used 120VAC. Now I remember. The rotor controller failed and the fail safe circuit didn't catch it. Burned a spot on the anode and gave us an instant "Gassed" tube. Started arcing. diposing of Possibly PCB Contaminate X-ray Tube Oil Your call. I suppose that a law breaking heathen *might* incinerated it by simply pouring it on a good lumber scrap fire although I personally would NEVER do such. :-) I have no doubt that burning will break down PCBs as incineration is how the EPA wants it disposed of, but I don't think I would want to stand around too close and breath the fumes. Getting rid of that stuff to EPA spec. is an absurd pain. There are only a few EPA certified incinerators in the country and It has to be sent to South Carolina from here to be disposed of in an EPA certified manner. Anyway,... tube? What tube? Sam who?... Don't recall meeting anyone by that name... Obtaining Used X-Ray Equipment (From: Kristian Ukkonen (kukkonen@cc.hut.fi).) You can find them from companies that sell x-ray equipment or overhaul them. They will often get old ones when they sell new ones that replace the old ones in hospitals etc.. Either they storage them as spare-parts, or sell them to junkyards. Anyway, call to all hospitals and companies that sell/overhaul x-ray equipment and ask for old ones. If you have a realistic need for one, you might get one. Talk to the engineers etc., not the byrocrats. Remember to emphasize that you need it as voltage source, NOT for producing x-rays (which requires permits etc.). If that doesn't work, ask what junkyards they sell the old ones to, and contact the junkyards and offer reasonable amount of money for one, so the junkyard will save one for you and call you to pick it up. The value of them for junkyards is the junk value of copper (about 2-4 usd/kg) in coils, the iron and transformer oil is practically worthless to them. It is usually a good idea to get to know your local junkyards. After you have been acquiring strange pieces of equipment from them for a while they will call you when interesting things arrive. :) I have so far found mass spectrometer, spectrofotometer, two sems, various high-vacuum pumps, x-ray transformers and tubes, hv-transformers etc. etc. from junkyards. They are a real goldmine for people who know what the junk is. It is absolutely amazing what research institutes, hospitals and universities throw away.. Shipping X-ray tubes (inserts): The only relatively common similar item I know of that is more fragile than a laser tube is a rotating anode X-ray tube. (This is what X-ray types call the "insert", not the entire X-ray head.) With these, the heavy anode/motor assembly - which may weigh several pounds - is attached to the glass envelope only at one end with most of the mass at the unsupported end. So, even though the glass is rather thick and would normally survive some trauma, a relatively modest physical shock will cause the tube to fracture. To have any chance of survival during shipping, the anode/motor assembly must either be secured to a rigid structure as it is when mounted in the X-ray head assembly so that it can't flex with respect to the glass envelope, or the entire glass tube must be packed with something like 12 inches of soft foam rubber all around to minimize the g-forces when the box drops onto the sorting conveyer from 10 feet up. And even this is no guarantee.) Anotomy of a rotating anode X-ray tube One doesn't find that many of these at garage sales but some do turn up.

Most higher energy X-ray devices (e.g., general radiology, CT scanners, etc.) use a rotating anode X-ray tube. The rotating anode is necessary to spread the extreme heat from the high energy electron beam - which account for 98 to 99 percent of the power dissipation - only around 1 percent ends up as X-rays. The anodes range from 2 or 3 inches to 7 inches or more in diameter and are made of tungsten, rhodium, and other exotic materials possibly backed by graphite. An induction motor built into the rotor assembly with the stator coil outside the glass envelope spins the anode at 3500 to 10,000 rpm. A tube like this would normally be encased inside an oil-filled housing and driven by 50 to 150 kV at up to several hundred mA. The filaments run on several volts at several amps. The temperature of the filament is what determines the current drawn at any given anode voltage. The split phase induction motor runs on 115 VAC and may require a small motor run capacitor.

A typical X-ray head or housing is shown in Bennet X-ray Machine Head Assembly. An X-ray tube and motor stator similar to the one inside this head is shown in GE X-ray Tube and Motor Assembly and removed in Typical GE X-ray Tube (Insert) - View 1 and Typical GE X-ray Tube (Insert) - View 2. A part of the anode can be seen through the circular window in the last photo.

Here are some photos of a rather large X-ray tube (close to CT-class) that didn't survive shipping: