Allen MOS-301 card reader woes

[Philip the organist]

Hi everyone,</p>

The card reader on our MOS301 at church has never worked properly. For some reason, whatever cards are inserted come out sounding like that card's voice combined with a Krummhorn. This renders the flute and percussion cards useless, and the reed cards greatly weakened and perverse-sounding.</p>

I specifically need the Festival Trumpet card for choir to work on Sunday, but right now it sounds like a Krumet. Could the problem be a dead bulb, resulting in the information from one row being lost? I've already wiggled the connectors on and off, and that didn't help at all. Is there a quick fix I could do without calling a tech? Is there a way that I could swap in another bulb, or one of a different type? In theory, wouldn't a large light mounted over the whole card assembly do the same thing? I know I'm inserting the cards the right way-text up, notched side in first. </p>

[^o)]</p>

[davidecasteel]

Re: Allen MOS-301 card reader woes



In a MOS instrument, looking into the card reader from the front of the organ one should see lamps lit in this pattern:</P>


O O O O O O O O O O (note the missing lamps in the 3rd and 11th positions from the left).</P>


Looking from the rear, the pattern would be reversed, of course. A burned-out lamp (or lamps) would definitely alter the nature of the voice produced by the card. If either of the left-most 2 lamps are not working, it would be impossible to read in a card at all (those are the clocking pulses). I'm not sure, but the right-most lamp may not be particularly critical (I think it just senses that a card has been inserted into the reader, and if the lamp is out it would just think there was a card there all the time). In the centralmain array of 7 lamps, those toward the left have the largest impact on the shape of the waveform; a failure ofa lampon the rightedge of that array might still permit the voice to be recognized; dead bulbs on the left of that array would totally destroy the nature of the voice.</P>


David</P>
<P mce_keep="true"></P>
<P mce_keep="true"></P>

[jbird604]

Re: Allen MOS-301 card reader woes



Philip,</P>


If there is anyone in the church who can use a soldering iron with care, he or she could change out a lamp. It is a 12 volt lamp and there used to be a suitable replacement at Radio Shack. If you see one is dead, clip it out with a wire cutter (or even finger nail clippers) and take it to Radio Shack for a match up. Just be sure to get a 12 volt lamp, as the power supply sends around 8 volts to the reader and a lamp rated for only 5 volts would quickly burn out.</P>


To remove the CR from the organ, you remove a couple of screws holding it in place (after pulling off the cable connector). This is not easy, but with care and patience you can get the reader out. Since it's not working you don't have much to lose.</P>


If you go to this much trouble, and if your makeshift tech is willing, change out the bulbs on either side of the one you're replacing because for some reason changing out a bulb often results in the adjoining ones going out. Some techs go so far as to change them all at once. Then test the reader before actually putting it back into the organ with screws. This will save a lot of frustration if another lamp turns out to be dead!</P>


Also, for the long term, there is a kit you can order from Harrison Labs that replaces all the lamps with LED's which will last for a lifetime. It's not a simple drop-in replacement for the lamps, though, requires a bit of drilling, etc. And a resistor is necessary in series with each LED. LED's cannot be directly substituted for lamps, unfortunately.</P>


You can also try using some canned air to thoroughly blow out the card reader. Debris in the holes can keep it from working even if all the lamps are burning.</P>


If you cannot get the CR repaired, combining the Krumhorn and the Schalmei on the great will give you a pretty strong trumpet, which you can beef up by adding the Prinzipal and the Hohlflote. Maybe not the Festival Trumpet tone you want, but quite a bit of brass in there. On some of the larger MOS organs, the Trumpet stop on the great was actually derived by drawing the Krumhorn and Schalmei together using a pair of diodes behind the tab.</P>


John</P>
<P mce_keep="true"></P>

[Philip the organist]

Re: Allen MOS-301 card reader woes



Yay!!! I am on the verge of a victory dance over here!! I had a rare hour or so alone with the organ before service tonight and I made very significant progress, thanks to David and John's advice.</p>

It turns out that lamp #5 is out (<span style="font-weight: bold;">OO O</span><span style="text-decoration: underline;"><span style="text-decoration: line-through;">o</span></span><span style="font-weight: bold;">00000 0</span>). I was able to get the reader out and clip off the bulb. I'll take it to Radio Shack tomorrow and see about a replacement. I have the card reader in my book bag so I'll have time to replace all the lamps before Sunday, and I can just hook it back up before church.
</p>

Even better news was that the combination action is now working!!![:'(] I figured that while I was poking around, I should take out the old lead-acid batteries to avoid leakage. Lo and behold, after powering on, it could set and remember generals and locals perfectly!!! </p>

It was indeed a very good evening!!</p>

Thank you again.
</p>

[davidecasteel]

Re: Allen MOS-301 card reader woes



Good job! I hope you can find a suitable replacement bulb and get it in place. For what it's worth, the Row 5 position corresponds to the MSB (Most Significant Bit) of the binary code (Row 6 is the sign bit) and thus has a value of 32; since the total range of values only is -64 to +63, having that bit missing has a huge effect on the values read in and the overall shape of the audio waveform. I think you will find a great improvement when the dead bulb is replaced. I'm also glad that I could help.</P>


David</P>

[jbird604]

Re: Allen MOS-301 card reader woes



Wonderful news! Like getting a new organ all at once!</P>


Now that the action works, you can go back to some of my previous advice about a battery backup, if you feel like you need that.</P>


Have a great day, and best wishes for the music this weekend.</P>


John</P>
<P mce_keep="true"></P>

[Philip the organist]

Re: Allen MOS-301 card reader woes



OK, I bought a replacement bulb at Radio Shack tonight and soldered it in myself! Fun! (I didn't burn myself this time, either!) I hooked a battery to the row of lights and all the joints are good and all lights came on. </p>

I'll report on Sunday...
</p>

[davidecasteel]

Re: Allen MOS-301 card reader woes



Great! Good luck! And do let us know how it went.</P>


David</P>

[jbird604]

Re: Allen MOS-301 card reader woes



Great! Now when you re-install the card reader, be absolutely sure to put the connector back on right. It will go on upside down as easily as the right way.</P>


Some card readers were marked with a yellow dot on both the reader itself and the cable connector. If yours is not, just look at the wire wraps. You'll notice that one side of the connector has only two (2) wires wrapped, and these are the two that provide the lamp power. Be sure the connector is put on so that those two wires are connecting to the traces that go directly to the lamps.</P>


(I believe I'm telling the truth here. I know this is the case withADC readers, and I think the MOS is wired the same way.)</P>


Also, before you install the reader you need tolook at the lamps to see if the one you installed has its filament (glowing hot thing) at about the same level physically as the rest. This will insure that this lamp actuates the photo-transistor on the other side about as well as the rest of them do. If the filament is way out of position that hole-row might not get read correctly. (You can attach the battery to make the lamps glow so you can see the filaments easily.)</P>


This is awesome, Philip. You will need to be a little bit savvy about technical stuff all your life if you are going to be around organs, either pipe or electronic, because SOMETHING is always going to be out of order on any organ you play! If you can fix little things you'll save a lot of money and you'll get to enjoy the instruments that much more.</P>


John</P>
<P mce_keep="true"></P>

[Philip the organist]

Re: Allen MOS-301 card reader woes



Hi guys,</p>

I got it in this morning. All the lamps lit up and the tone quality sounded right. Unfortunately the volume of all the cards is unusually low for some reason-even softer than the Flauto Dolce on the Great, rendering the card reader feature next to useless. Is there a gain control that I could adjust to boost the volume to a proper level?
</p>

[davidecasteel]

Re: Allen MOS-301 card reader woes



Philip, I'm sorry but I don't have any knowledge about the reader itself, or how it fits into the organ system (just the coding). One of the other threads had a comment about some ADC organs being able to use the MOS cards but that the volume would be wrong; however, since this is a MOS instrument that should not be a factor. If all the cards are too quiet, it would seem that some volume adjustment must be responsible and therefore capable of being remedied. Hopefully, someone else following this thread who is more cognizant of the way the reader actually works will post an answer for you.</P>


I suppose it is possible that all the cards you have are for soft stops, but that seems unlikely. If you can scan one of your cards (the one you think should be loudest would be best) and send me the image, I'll analyze it and see if it is coded for max volume.</P>


David</P>

[jbird604]

Re: Allen MOS-301 card reader woes



Philip,</P>


There is no volume adjustment for the cards (alterable voices) in MOS. Alterable voices are generated right on the main MOS board in the same manner as the resident voices, not on a satellite board, as in ADC and other systems.</P>


Perhaps one or more of the light holes in the reader has some fuzz or other debris in it. I know you don't relish the idea of taking the CR out again, but I think you must. You can dismantle the entire thing very carefully... take note of every screw, nut and bolt, how the spacing washers go in, etc. Be sure you can put it back together just as it was.</P>


While you have it dismantled, use canned air to blow out the slots between the lamps and the photo-transistors on the receiver board. Be sure there is no dust, fuzz, paper scrap, or other stuff in the card slot or anywhere else. Be sure the photo transistors are clean.. blow them with the air, wipe gentle with Q-tips, etc. </P>


Also be sure the lamp filaments all line up and are down in the holder as far as they need to be. As a last resort, you can boost the card reader voltage by a half volt or so. You will find an adjustment for that voltage on one of the power supplies. If you check it with a meter, it should read around 8 volts, but it won't hurt anything to set it as high as 9 or 9.5 volts, if that will make the reader work.</P>


David might be able to tell you which (if any) of the hole-rows decode the intensity or volume level of each card. If he can, you could concentrate on making sure that hole is clean and the lamp properly positioned before going to all this trouble.</P>


John</P>
<P mce_keep="true"></P>

[davidecasteel]

Re: Allen MOS-301 card reader woes



As I understand the coding on the cards for MOS, there is no provision for a "scaling" factor that adjusts the volume (intensity) of the generated sound. Instead, the binary logic uses the data in the 16 active columns to specify the shape of the waveform and its relative amplitude. The 16 points may have values between -64 and +63, and if plotted against time those points will trace the outline of a half-cycle of the waveform (what you'd see on an oscilloscope). The other half of the waveform is created by point-symmetry by inverting the values and their positions in time (reverse order). The resultant waveform is then replicated over and over at the rate determined by the pitch of the key depressed. The loudest stop possible would be a square wave alternating in value between +63 and -64 (or -63, probably, to be symmetrical). A card producing this voice would have all 16 columns coded as +63 (and the other half would then be -63). This would be a very loud sound sort of like a raucous oboe. Any useful voice will be less loud than that, of course. It is possible to create voices that have identical shapes of their waveforms (same relations among the harmonics) but different amplitudes--cutting the values in half will decrease the sound by 6dB (I think I have that right--it's a voltage relationship). So, in reality, the relative amplitudes of the card voices are determined by the set of 16 values encoded on the cards--louder voices have more points at greater amplitude values, and softer ones have smaller values. To be accurate, the loudness is more complicated than that, as the area under the curve is probably just as important as the maximum amplitude, but I'm not good enough at this business to know all about that.</P>


The replication of the waveform cycle so specified is controlled by the frequencies of the 8' register--pressing middle C will replicate the pattern 261.7 times every second the key is pressed, and the A above will do it 440 times per second, etc. There is no provision (that I know of) for changing the impressed frequency to a different register (using twice the frequencies would scan the waveform at a 4' register rate, for example). However, it is possible to encode 4' stops just by encoding a waveform that has a complete cycle defined by the 16 points (say by using the first 8 points to define the positive half-cycle and the other 8 to define the negative half-cycle; when the circuitry does the point-symmetry process it thus produces 2 complete cycles during the 8' register repetition period, and thus creates a tone an octave higher. My spreadsheet handles this automatically when the harmonics are selected--beginning with 4' pitch (2nd harmonic) and using higher harmonics will produce a 4' stop. Although it is possible toproduce patterns that are not symmetrical in the half-cycle at the higher pitches,it is not easy to predict what the effect will be. If a waveform half-cycle specification (16 points) lies completely on one side of the time axis (all positive or all negative), the tone produced will definitely be some kind of 8' stop; if it lies on both sides of the time axis, it will have some strong content of the higher harmonics, even to the extent of overriding the fundamental.</P>


When I first developed my method for analyzing and producing new tone cards, one of my major functions was to "rescale" voices that our organist liked but found to be too loud or too soft for her needs. In every case where I produced new cards with all values adjusted by the same factor, the overall sound of the voice was the same as the original, but louder or softer (as intended). This was one way I was fairly sure that my analysis process was correct.</P>


If the card reader holes are plugged and not allowing the punched holes to be read correctly, I would expect that the voice itself would be seriously compromised due to the distortion of the waveform caused by missing bit values in the binary code. Philip has indicated that the voices sound OK, which I interpret that a flute stop sounds like a flute, an oboe like an oboe, etc. Since the code is digital, a bit is either detected or it is not, and missing bits can have drastic effects on the nature of the voice (and, depending on which bits they are, also on the volume).</P>


In the above, I have been discussing the function of the holes punched in rows 0 through 6, which is where the binary code is located. The other active rows contain punches in rows 8 and 9, and there is a pair of these punches for each of the 16data columns. The function of these punches is to gate the binary value represented by the punches in rows 0 through 6 into the array of values used by thedigital-to-analogconverters in the tone generators. (I am thinking of the process as taking place in the analog realm; however, for all I know, it is done in binary form by being mixed with the other binary data from the fixed stops, and then converted to analog before amplification.) My ignorance of the actual process taking place is abysmal. I think the card reader also has a lamp and sensor for the 12 row (top row) and I really do not know what the function of that sensor is; many cards I have seen do not have a punch anywhere in that row; certainly, none of my special cards did. I have assumed that it is used principally to let the reader know when a card has been inserted--maybe when the card is inserted the data array in the reader is cleared, and when the card is withdrawn, the sequence of punches in the 8 and 9 rows triggers placement of the 16 values into the array, and when the card is removed those values are transferred to the Alterable Voice registers. It's a plausable process, but couldbe totally fallacious.</P>


I welcome further enlightenment from anyone on the operation of this equipment. I've told you all I could deduce about how it works, code-wise. The discussion by another member on another thread about his modification to enable direct entry of data values did go into the way the 8 and 9 row punches controlled the data input.</P>


David</P>

[Philip the organist]

Re: Allen MOS-301 card reader woes



Thanks John and David. That was some fascinating information about the technology!</p>

The voices (best I could tell over the chatter of a hundred early parishioners or so), though very quiet, sounded about right-some digital noise to it-but you could still tell the Rohrflote from the Gedackt card and so forth. The Festival Trumpet card sounded milder and thinner than it should have, and was very quiet, so I guess there must be some sort of detection error. I'll try boosting the voltage, and if that doesn't work, I'll take it out again and clean it per John's instructions. </p>

Thank you again, very, very much.</p>

PF
</p>

[jbird604]

Re: Allen MOS-301 card reader woes



David,</P>


I've often wondered why the extra holes at the back end of the card (where you hold the card while inserting it). These holes never even enter the reader, so they can't have any voice data in them. Do you know what they represent?</P>


Philip -- Before you replaced the lamp, what did you get from the alterables? I think you said it was a reed-like tone, but I'm wondering about the volume level. Was it louder than the more correct tones that you are getting now? Or was it also on the soft side?</P>


I'm still holding out hope that you can repair the reader by very carefully cleaning it up and ailgning the lamps. But there is a chance the problem is in the MOS board itself or possibly in the card reader logic circuitry contained on the clock board.</P>


We are heading to a church this very day to (maybe) install our first Harrison Labs LED kit to replace the card reader lamps. This is a church 2+ hours away and they have had card reader lamps go out at least a couple of times in the last year, so we're hoping to solve the problem once and for all with the LED kit. Will report on that in a new thread if we do it today.</P>


If it's as easy as the instructions make it outto be, this could be the way to go for you.</P>


John</P>