Hi, I recently got a Sulzer 5A from a friend. It is mounted on a plate that includes an oblong box mounted on each side. The box on the left divides the 100 Khz output down to 0.1 PPS. The box on the right has a few components in it, but mostly switches the battery from Internal to External, and has the 0.1 PPS output connector and such. Pictures attached. The divider is a most unusual design (to me). It has a series of modules, most of which are divide-by-5, but the *power* for the modules are wired in an odd series/parallel combination. That is, the +V goes to the last divider in the chain, and that module's "-V" goes to the +V of the next module. But then two pairs of modules have their +V and -V wired in parallel with each other. Each divide-by-5 includes five 2N706's, with 3 diodes on the Base of each Q, etc. Also, the back end of the divider box has two buttons: One doubles the 0.1 PPS to 0.2 PPS; the other halves it to 0.05 PPS. They have Dymo labels for FAST and SLOW. Does anyone recognize this type of circuit? Are the two accessory boxes strictly one-off, or are they a known thing? Thanks for any comments! Pete

Peter, Several time nuts have examples of the Sulzer 2.5 or 5A oscillator. Yes, it is typically the dual cylinder configuration, one for the ovenized oscillator and one for the power supply / backup battery. See pages such as: (1) https://geekblog.febo.com/?p=17 (2) https://febo.com/for_sale/sulzer/ (3) http://leapsecond.com/museum/sul25-1/ (4) http://leapsecond.com/museum/sul25-2/ Note that link (4) include a schematic for the power supply module and (3) has a schematic for the double oven oven oscillator and frequency divider. Now, about your photo. To me, the shape, placement and dymo label suggests that it is a non-standard DIY add-on, in which case you're on your own to trace the schematic and get it working. However, it may borrow from the original Sulzer divider circuit design as seen in (3). I can't answer your +V and -V question. If this is a regenerative divider the design may include fail-safe operation, that is, if there is a power glitch or missing input cycles the divider automatically stops and must be manually restarted. This prevents accidental time slippage. Perhaps post photos or reverse engineered schematic of your divider module. One of the vintage / RF members on the mailing list might immediately recognize the design. /tvb

-------- Tom Van Baak via time-nuts writes: > I can't answer your +V and -V question. I have seen similar kind of "power-saving" in other contexts, where a battery supply was much higher voltage than the secondary circuits which needed power. Instead of dumping a lot of power in a linear regulator for each secondary circuit, the become each others pre-regulators. To make it work, they need to consume approx the same current, which is almost certainly why two of your dividers are in parallel. There are a LOT of reasons not to do things that way, but power- (or heat-) considerations can override them. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.

Hi I very much agree with PHK on this one. That’s both for the “why” and very much for the “why not” do this. If you dig into various early solid state portable transceivers, you will find them doing this sort of “stage stacking” in a number of places. You got more battery life as a result. You also got some *really* weird troubleshooting issues. Best guess here is that the Sulzer didn’t have a lot of “spare current” out of the supply. Who ever designed that add on did what they had to do, Bob > On Jan 7, 2025, at 8:13 AM, Poul-Henning Kamp via time-nuts <time-nuts@lists.febo.com> wrote: > > -------- > Tom Van Baak via time-nuts writes: > >> I can't answer your +V and -V question. > > I have seen similar kind of "power-saving" in other contexts, where > a battery supply was much higher voltage than the secondary circuits > which needed power. > > Instead of dumping a lot of power in a linear regulator for each secondary > circuit, the become each others pre-regulators. > > To make it work, they need to consume approx the same current, which is > almost certainly why two of your dividers are in parallel. > > There are a LOT of reasons not to do things that way, but power- (or heat-) > considerations can override them. > > -- > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 > phk@FreeBSD.ORG | TCP/IP since RFC 956 > FreeBSD committer | BSD since 4.3-tahoe > Never attribute to malice what can adequately be explained by incompetence. > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com

Hi Tom and Poul-Henning, Here is a partial sketch of one of the divide-by-5 modules. Each Q has 3 diodes, and they are interconnected with other stages in various ways to achieve the divide-by-5 function. I may have drawn the diodes backwards, but you get the idea - it's similar to DTL gates and other circuits used in early computers. The dividers are self-starting, and there is no signal to them other than V+, V-, IN, and OUT. Note how the outputs are C-coupled, and all the Q's are in parallel to drive the overall output. I should have sketched more of the circuit before I reassembled it... It works, after fixing an open solder joint in the middle of the divider chain. Tom: Thanks for those links! I had found a complete manual online for the 5A (and 2.5). The overall unit works, but I don't have the equipment or knowledge to understand "how well" it works. EXPERIMENT: I let it run for about 48 hours, then did a Lissajous comparison with my scope, against the output of a CW12-TIM GPSDO module. Inside the house, the CW12 LED-blinking pattern reported only one "locked" sat, and several others "visible" (does that mean that the DO output would be "good" for this type of experiment?? Or does it need multiple "locked" sats?? I understand that more locked sats are needed for 3-D positioning, but what about just the D.O. output?). The Sulzer reported correct numbers (matching the old markings on the unit) for the oven current readings, implying (I assume) that there are no big issues with the oven controls. The Lissajous would drift very slowly, with a period of a minute or more, and would sometimes change direction, and would sometimes "breathe" (oscillate at a somewhat faster rate, but on average stay at about the same point). I could tweak the Sulzer with it's fine adjustment, but after a couple minutes it would always drift slightly relative to the GPSDO. I would welcome comments on these experiment results! Pete On Tue, Jan 7, 2025 at 6:06 AM Tom Van Baak via time-nuts < time-nuts@lists.febo.com> wrote: > Peter, > > Several time nuts have examples of the Sulzer 2.5 or 5A oscillator. Yes, > it is typically the dual cylinder configuration, one for the ovenized > oscillator and one for the power supply / backup battery. See pages such > as: > > (1) https://geekblog.febo.com/?p=17 > > (2) https://febo.com/for_sale/sulzer/ > > (3) http://leapsecond.com/museum/sul25-1/ > > (4) http://leapsecond.com/museum/sul25-2/ > > Note that link (4) include a schematic for the power supply module and > (3) has a schematic for the double oven oven oscillator and frequency > divider. > > Now, about your photo. To me, the shape, placement and dymo label > suggests that it is a non-standard DIY add-on, in which case you're on > your own to trace the schematic and get it working. However, it may > borrow from the original Sulzer divider circuit design as seen in (3). > > I can't answer your +V and -V question. If this is a regenerative > divider the design may include fail-safe operation, that is, if there is > a power glitch or missing input cycles the divider automatically stops > and must be manually restarted. This prevents accidental time slippage. > > Perhaps post photos or reverse engineered schematic of your divider > module. One of the vintage / RF members on the mailing list might > immediately recognize the design. > > /tvb > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com >

Hi To get accurate time off of a single sat with a GPS module, you need an accurate location “locked in” to the module. Without that, there really isn’t a way to work out what’s what. Typically you run the module for a while to come up with that location. How long “a while” is depends a lot on how often you get more than 4 sats locked and how long they stay locked. Something like 48 hours of “locked at least 20 minutes” is a pretty good goal. Bob > On Jan 7, 2025, at 10:18 AM, Peter McCollum via time-nuts <time-nuts@lists.febo.com> wrote: > > Hi Tom and Poul-Henning, > Here is a partial sketch of one of the divide-by-5 modules. Each Q has 3 > diodes, and they are interconnected with other stages in various ways to > achieve the divide-by-5 function. I may have drawn the diodes backwards, > but you get the idea - it's similar to DTL gates and other circuits used in > early computers. > The dividers are self-starting, and there is no signal to them other than > V+, V-, IN, and OUT. > Note how the outputs are C-coupled, and all the Q's are in parallel to > drive the overall output. I should have sketched more of the circuit before > I reassembled it... > > It works, after fixing an open solder joint in the middle of the divider > chain. > > Tom: Thanks for those links! I had found a complete manual online for the > 5A (and 2.5). > > The overall unit works, but I don't have the equipment or knowledge to > understand "how well" it works. > EXPERIMENT: > I let it run for about 48 hours, then did a Lissajous comparison with my > scope, against the output of a CW12-TIM GPSDO module. Inside the house, the > CW12 LED-blinking pattern reported only one "locked" sat, and several > others "visible" (does that mean that the DO output would be "good" for > this type of experiment?? Or does it need multiple "locked" sats?? I > understand that more locked sats are needed for 3-D positioning, but what > about just the D.O. output?). > The Sulzer reported correct numbers (matching the old markings on the unit) > for the oven current readings, implying (I assume) that there are no big > issues with the oven controls. > The Lissajous would drift very slowly, with a period of a minute or more, > and would sometimes change direction, and would sometimes "breathe" > (oscillate at a somewhat faster rate, but on average stay at about the same > point). > I could tweak the Sulzer with it's fine adjustment, but after a couple > minutes it would always drift slightly relative to the GPSDO. > I would welcome comments on these experiment results! > > Pete > > > On Tue, Jan 7, 2025 at 6:06 AM Tom Van Baak via time-nuts < > time-nuts@lists.febo.com> wrote: > >> Peter, >> >> Several time nuts have examples of the Sulzer 2.5 or 5A oscillator. Yes, >> it is typically the dual cylinder configuration, one for the ovenized >> oscillator and one for the power supply / backup battery. See pages such >> as: >> >> (1) https://geekblog.febo.com/?p=17 >> >> (2) https://febo.com/for_sale/sulzer/ >> >> (3) http://leapsecond.com/museum/sul25-1/ >> >> (4) http://leapsecond.com/museum/sul25-2/ >> >> Note that link (4) include a schematic for the power supply module and >> (3) has a schematic for the double oven oven oscillator and frequency >> divider. >> >> Now, about your photo. To me, the shape, placement and dymo label >> suggests that it is a non-standard DIY add-on, in which case you're on >> your own to trace the schematic and get it working. However, it may >> borrow from the original Sulzer divider circuit design as seen in (3). >> >> I can't answer your +V and -V question. If this is a regenerative >> divider the design may include fail-safe operation, that is, if there is >> a power glitch or missing input cycles the divider automatically stops >> and must be manually restarted. This prevents accidental time slippage. >> >> Perhaps post photos or reverse engineered schematic of your divider >> module. One of the vintage / RF members on the mailing list might >> immediately recognize the design. >> >> /tvb >> _______________________________________________ >> time-nuts mailing list -- time-nuts@lists.febo.com >> To unsubscribe send an email to time-nuts-leave@lists.febo.com >> > <div.JPG>_______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com

This site has interesting pictures of a disassembled Sulzer 5B, as well as the AN/URQ-10. Unfortunately it only shows the thumbnails, as the resource directory with the larger images and PDFs is gone. http://home.catv.ne.jp/ff/y226/1/1-12/OCXO/sub1-12-OCXO.htm#5B Walt -----Original Message----- From: Tom Van Baak via time-nuts <time-nuts@lists.febo.com> Sent: Tuesday, January 7, 2025 5:04 AM To: time-nuts@lists.febo.com Cc: Tom Van Baak <tvb@LeapSecond.com> Subject: [time-nuts] Re: Sulzer 5A accessory divider Peter, Several time nuts have examples of the Sulzer 2.5 or 5A oscillator. Yes, it is typically the dual cylinder configuration, one for the ovenized oscillator and one for the power supply / backup battery. See pages such as: (1) https://geekblog.febo.com/?p=17 (2) https://febo.com/for_sale/sulzer/ (3) http://leapsecond.com/museum/sul25-1/ (4) http://leapsecond.com/museum/sul25-2/ Note that link (4) include a schematic for the power supply module and (3) has a schematic for the double oven oven oscillator and frequency divider. Now, about your photo. To me, the shape, placement and dymo label suggests that it is a non-standard DIY add-on, in which case you're on your own to trace the schematic and get it working. However, it may borrow from the original Sulzer divider circuit design as seen in (3). I can't answer your +V and -V question. If this is a regenerative divider the design may include fail-safe operation, that is, if there is a power glitch or missing input cycles the divider automatically stops and must be manually restarted. This prevents accidental time slippage. Perhaps post photos or reverse engineered schematic of your divider module. One of the vintage / RF members on the mailing list might immediately recognize the design. /tvb _______________________________________________ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe send an email to time-nuts-leave@lists.febo.com

Hi Pete: Here's some patents related to the Sulzer: https://prc68.com/I/timefreq.shtml#Sulzer Have you confirmed that the modules are divide by 5 rather than divide by 10? I ask because a four stage counter can work as a divide by ten, see: HP AC-4A Decade Counter https://prc68.com/I/HPac4a.shtml A five stage Johnson Twisted Ring counter is a way to get the highest counting rate from a flip-flop. https://prc68.com/I/comp.shtml#Disp -- Have Fun, Brooke Clarke https://www.PRC68.com axioms: 1. The extent to which you can fix or improve something will be limited by how well you understand how it works. 2. Everybody, with no exceptions, holds false beliefs. -------- Original Message -------- > Hi, I recently got a Sulzer 5A from a friend. > It is mounted on a plate that includes an oblong box mounted on each side. > The box on the left divides the 100 Khz output down to 0.1 PPS. The box on > the right has a few components in it, but mostly switches the battery from > Internal to External, and has the 0.1 PPS output connector and such. > Pictures attached. > > The divider is a most unusual design (to me). It has a series of modules, > most of which are divide-by-5, but the *power* for the modules are wired in > an odd series/parallel combination. That is, the +V goes to the last > divider in the chain, and that module's "-V" goes to the +V of the next > module. But then two pairs of modules have their +V and -V wired in > parallel with each other. Each divide-by-5 includes five 2N706's, with 3 > diodes on the Base of each Q, etc. > Also, the back end of the divider box has two buttons: One doubles the 0.1 > PPS to 0.2 PPS; the other halves it to 0.05 PPS. They have Dymo labels for > FAST and SLOW. > > Does anyone recognize this type of circuit? > Are the two accessory boxes strictly one-off, or are they a known thing? > Thanks for any comments! > Pete > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com