Not me. My experience with surfboards ended at about age 18 when I was knocked out by someone else's board and pulled out of the water by my friend at what is now Haulover inlet / beach. I probably logged thousands of hours in the water just North of Port Everglades and in the gulf, but I was on a 15 foot AMF Venture, or a 14 foot Hobie Cat. Both are one man sailboats that can fly! I also spent a good deal of time on some larger 34 and 42 foot sailboats in the Atlantic around South Florida and the Western Bahamas.is that George with the blond hair on that yellow board?
Ouch!!My experience with surfboards ended at about age 18 when I was knocked out by someone else's board
I used to windsurf and sail catamarans on Long Island Sound on the weekends in CT many many moons ago. I had a Hobie 14' and then an AMF Trac 16 catamaran. I still have the Trac 16 here near Lake Hickory, NC, but the wind is too sporadic to bother setting it up. A Yamaha VX110 Wave Runner with a start button and reverse lever fills the bill now. I still do some mild whitewater kayaking on the rivers near the mountains with a group here and for the last 20 years, but we're all in our 70s now. Some wild times, some I wouldn't want to repeat, just want to remember.
We (Kathy was driving) saw one of these hydroplaning stand up boards on Lake James a month ago ( like in the video but flat water! ). Looked like the guy was standing on a flying carpet at first. It must have been the version with an electric motor/prop on the hydroplane. Those can do 25 mph with Lithium batteries. All the boaters were watching him zip around. I think the ones in the video may have been the non-motorized ones, you have to pump the board up and down. 10 to 1 price difference unfortunately. Sure looks like fun, but need to shed 50 years off to balance like that.
I was thinking to buy the utracer, but having a scope the Tek 576 seems a better solution. What do I have to look at when buying it?Depends on where the grid1 voltage is nominally biased. (the Plate Fdbk resistor is varying that around positively too) The cathode only needs to come up to the max pos. grid1 voltage.
I would like to test the pentode+pmosfet as a single triode, so driving the gate of the pmosfet and ground its drain and the feedback voltage divider.
We (Kathy was driving) saw one of these hydroplaning stand up boards on Lake James a month ago
I saw them in Italy at Garda Lake too.
They do look funny, but I’m pretty sure no one was older than 25 on the non-electric ones. Electric seems amazing and more ageing-friendly.
Last edited:
Local direct feedback dc coupled to the anode of the output tube and the secondary of the OPT instead of ground.
I was thinking about this solution. I’m not in Italy, so just thinking loud, but seems reasonable: connecting the voltage divider (direct local feeeback) a-g1 to anode and secondary of the OPT instead of ground has a positive and a negative effect.
Both sides of the voltage divider see a swing that is is phase. This means that g1 reaches 0 V when the anode is at an higher voltage compared to when the bottom of the voltage divider is grounded.
This has an advantage: the pmosfet can work at a lower idle dissipation compared to the same amount of feedback given by increasing the voltage divider ratio. Seen from another point of view: you get better THD results (including the OPT) with the same pmosfet idle dissipation.
And a disadvantage: the anode cannot swing down as before, so the maximum power of the amp is lower.
@Tubelab_com
Have you tried those as cathode drivers in the driver stage?
https://www.onsemi.com/pdf/datasheet/j175-d.pdf
Have you tried those as cathode drivers in the driver stage?
https://www.onsemi.com/pdf/datasheet/j175-d.pdf
If the uTracer or eTracer lets you select the negative stepping range, you can always just put a floating linear supply in series (adjustable supply best) to get the correct positive range for driving the Mosfet gate. No doubt those will be simpler than getting a working Tek 576 running. You probably can't drive the cathode directly with those however.
The Tek 576 models on Ebay are typically quite expensive for one that is guaranteed to be working. ( $3000-$4000) I have gotten 3 now that were in the $400 to $600 range and just fixed the problems. Some problems need to be avoided however. You want one that can at least put some luminous streaks or dots on the screen. Otherwise the CRT or HV supply (ferrite HV xfmr) may be shot. However, I've seen a used working CRT on Ebay for around $350. And I have a spare working CRT here, latest rev. 03 version, but with a small screen/phosphor burn spot at the 0,0 origin, you can have it for free if needed. The burn spot just nulls out the annoying bright spot when not actively tracing. (almost a plus ) The HV supply for the CRT is available on Ebay for around $150:
https://www.ebay.com/itm/145443118170 (not guaranteed working however)
The other common problem with these used units are the push-button switches on the front, particularly the two that are marked "Invert". These get pressed every time the unit gets flipped from NPN to PNP (etc), so get a lot of wear. A bit of work to get to them, but only the plunger part needs replacing (no un-soldering the 50 PCB terminals! ) New Centralab switches can be found to remove the plungers from for substitution.
Another odd problem is the plug-in relay that operates the HV Collector sweep transformer. The 576 comes new with a plastic cover that flips down over the transistor test terminals for HV safety. This usually gets discarded or lost by the time it has reached Ebay. The cover activates the relay when closed so HV collector/drain voltage can be applied. Users seem to stick a pencil in the contact hole to get it to over-ride the safety. This pencil cheat scheme seems to be generally flaky and the relay chatters on and off until its contacts burn off. Easy to fix the plug-in relay and add a toggle switch on the test unit to over-ride (permanently for tubes).
The test terminal box in front has a toggle switch lever for flipping between the two test device positions. This switch is often broken off from poor handling or packaging, but can be fixed with a new SwitchCraft switch.
For pentodes, you need to add some provision to dis-connect the screen V from the tube when not actively tracing. I used one of the "sense" V contact sets on the toggle switch to connect (individual in my setup) screen V supplies to each tube when tracing is selected.
These Tek 576 units are from 1968 up thru 1988 I think. The older units are reaching the limit for the electrolytic caps (mainly power supply). So these may need changing, particularly for the lower serial number units. If you can find a 576 with the later IEC AC cord plug-in connector on the back, you will have one of the later units that are likely still in good shape for electrolytics and possibly the "Invert" switches.
You will want a Tek service manual for the 576. These are like 1 1/2 inches thick, or on CDs. The schematics fold out, so the real manual is easier to use.
The PC boards do not plug in generally with edge connectors, but rather have individual push-on wire connectors. Make sure to write down the wire colors if removing a PC board, although the manual has annotated pictures too. The manual has a good circuit description write-up for how it works as well as servicing info and parts.
The power supplies are all regulated, but they key off one supply for voltage reference. The voltages are all on one PC board edge, so that is the 1st place to start looking for any problem.
Oh, and these things weigh around 75 Lbs before packaging or palleting for shipping. Can be costly to ship if not nearby. Tell them to pack it well, especially the front panel.
The Tek 576 models on Ebay are typically quite expensive for one that is guaranteed to be working. ( $3000-$4000) I have gotten 3 now that were in the $400 to $600 range and just fixed the problems. Some problems need to be avoided however. You want one that can at least put some luminous streaks or dots on the screen. Otherwise the CRT or HV supply (ferrite HV xfmr) may be shot. However, I've seen a used working CRT on Ebay for around $350. And I have a spare working CRT here, latest rev. 03 version, but with a small screen/phosphor burn spot at the 0,0 origin, you can have it for free if needed. The burn spot just nulls out the annoying bright spot when not actively tracing. (almost a plus ) The HV supply for the CRT is available on Ebay for around $150:
https://www.ebay.com/itm/145443118170 (not guaranteed working however)
The other common problem with these used units are the push-button switches on the front, particularly the two that are marked "Invert". These get pressed every time the unit gets flipped from NPN to PNP (etc), so get a lot of wear. A bit of work to get to them, but only the plunger part needs replacing (no un-soldering the 50 PCB terminals! ) New Centralab switches can be found to remove the plungers from for substitution.
Another odd problem is the plug-in relay that operates the HV Collector sweep transformer. The 576 comes new with a plastic cover that flips down over the transistor test terminals for HV safety. This usually gets discarded or lost by the time it has reached Ebay. The cover activates the relay when closed so HV collector/drain voltage can be applied. Users seem to stick a pencil in the contact hole to get it to over-ride the safety. This pencil cheat scheme seems to be generally flaky and the relay chatters on and off until its contacts burn off. Easy to fix the plug-in relay and add a toggle switch on the test unit to over-ride (permanently for tubes).
The test terminal box in front has a toggle switch lever for flipping between the two test device positions. This switch is often broken off from poor handling or packaging, but can be fixed with a new SwitchCraft switch.
For pentodes, you need to add some provision to dis-connect the screen V from the tube when not actively tracing. I used one of the "sense" V contact sets on the toggle switch to connect (individual in my setup) screen V supplies to each tube when tracing is selected.
These Tek 576 units are from 1968 up thru 1988 I think. The older units are reaching the limit for the electrolytic caps (mainly power supply). So these may need changing, particularly for the lower serial number units. If you can find a 576 with the later IEC AC cord plug-in connector on the back, you will have one of the later units that are likely still in good shape for electrolytics and possibly the "Invert" switches.
You will want a Tek service manual for the 576. These are like 1 1/2 inches thick, or on CDs. The schematics fold out, so the real manual is easier to use.
The PC boards do not plug in generally with edge connectors, but rather have individual push-on wire connectors. Make sure to write down the wire colors if removing a PC board, although the manual has annotated pictures too. The manual has a good circuit description write-up for how it works as well as servicing info and parts.
The power supplies are all regulated, but they key off one supply for voltage reference. The voltages are all on one PC board edge, so that is the 1st place to start looking for any problem.
Oh, and these things weigh around 75 Lbs before packaging or palleting for shipping. Can be costly to ship if not nearby. Tell them to pack it well, especially the front panel.
Unfortunately, the status of all Sony Tektronix 370B curve tracers are subject to one or more of the following:
Un-Obtan-ium
Ob-solete-ium
Over-priced-ium
And locked up in an underground bunker, with 24-7 multiple pairs of armed guards.
Oh how I miss all the really good kinds of test equipment that I used to operate and support.
Me . . . yes, Spoiled Rotten.
Un-Obtan-ium
Ob-solete-ium
Over-priced-ium
And locked up in an underground bunker, with 24-7 multiple pairs of armed guards.
Oh how I miss all the really good kinds of test equipment that I used to operate and support.
Me . . . yes, Spoiled Rotten.
At this point I have never tried any J-fet for the cathode driver. I did find three 2N5461's in my parts bin. I'll try them when I get time to experiment with the driver stage.@Tubelab_com
Have you tried those as cathode drivers in the driver stage?
Here is a Tek 370B curve tracer for just $18000
https://www.ebay.com/itm/354770788712
and the Tek 371B curve tracer for a steal at $25000 plus $1000 shipping
https://www.ebay.com/itm/233848794950
well, I did see a Tek 370 for $3200 but it has some display glitch. Some cheaper 370B's listed too.
The Tek 576 is looking pretty good at $400 despite not having a floppy disk or IEEE 488 port.
Some other issues with the Tek 576 I thought of:
Make sure the Collector Supply selector knob is not cracked or missing. It has markings on a clear plastic skirt. An Un-obtanium knob. I guess one could make a new clear plastic skirt and glue it onto a plain knob.
The max collector sweep V selectible with that knob is 1500V. I modified the HV rectifier setup so its now 750V max at twice the current. A little more useful for Sweep tubes etc and safer for me.
The Horizontal V/div switch assembly has a variable cap on it. I was getting crazy results until I blew the dust out of it.
There is a simple mod that can be done to the Step Generator board to get a max of 15 traces instead of 11. You'll notice the extra traces on my later graphs.
I added a pot and knob on the side panel to provide a continuously variable step size, selectible by a toggle switch. The standard 1-2-5-10... step selector switch doesn't always use the most screen area for some tubes. Better graphs available with this, but you have to check the actual step size. There's a switch position to do that already, no big deal.
I modified the step amplifier output stage with higher V transistors so it could handle a +/-200V max step range (at up to 2A). Standard 576 only does up to +/-20V, not enough for tubes. And I added a boost supply for the new step amplifier which allows up to +/-170V range (using an off the shelf isolation xfmr. But only good for 1/2 an Amp step current. An External boost supply needed for full 2 Amp stepping. Can do grid2 curve sets with this now. (one can also add a series DC supply to the screen pin to do a higher screen V range, like 50V to 220V, or 100V to 270V)
That Step Amplifier module mentioned earlier can be used instead for getting up to -200V step range for tube (neg. 200v only), with no 576 mods, but only up to 20 mA grid current. (ie, limited cathode drive )
https://www.ebay.com/itm/354770788712
and the Tek 371B curve tracer for a steal at $25000 plus $1000 shipping
https://www.ebay.com/itm/233848794950
well, I did see a Tek 370 for $3200 but it has some display glitch. Some cheaper 370B's listed too.
The Tek 576 is looking pretty good at $400 despite not having a floppy disk or IEEE 488 port.
Some other issues with the Tek 576 I thought of:
Make sure the Collector Supply selector knob is not cracked or missing. It has markings on a clear plastic skirt. An Un-obtanium knob. I guess one could make a new clear plastic skirt and glue it onto a plain knob.
The max collector sweep V selectible with that knob is 1500V. I modified the HV rectifier setup so its now 750V max at twice the current. A little more useful for Sweep tubes etc and safer for me.
The Horizontal V/div switch assembly has a variable cap on it. I was getting crazy results until I blew the dust out of it.
There is a simple mod that can be done to the Step Generator board to get a max of 15 traces instead of 11. You'll notice the extra traces on my later graphs.
I added a pot and knob on the side panel to provide a continuously variable step size, selectible by a toggle switch. The standard 1-2-5-10... step selector switch doesn't always use the most screen area for some tubes. Better graphs available with this, but you have to check the actual step size. There's a switch position to do that already, no big deal.
I modified the step amplifier output stage with higher V transistors so it could handle a +/-200V max step range (at up to 2A). Standard 576 only does up to +/-20V, not enough for tubes. And I added a boost supply for the new step amplifier which allows up to +/-170V range (using an off the shelf isolation xfmr. But only good for 1/2 an Amp step current. An External boost supply needed for full 2 Amp stepping. Can do grid2 curve sets with this now. (one can also add a series DC supply to the screen pin to do a higher screen V range, like 50V to 220V, or 100V to 270V)
That Step Amplifier module mentioned earlier can be used instead for getting up to -200V step range for tube (neg. 200v only), with no 576 mods, but only up to 20 mA grid current. (ie, limited cathode drive )
Last edited:
smoking-amp,
The real cost of a 370B is when repairs are needed.
Time to search for parts; price of parts; having to re-design a circuit due to a lack of a needed part.
To test tubes on a 370B, an amplifier is needed to increase the negative offset voltage, and negative step voltages for testing higher power tubes.
One thing I loved about the 370B: put a dot on a curve, move the dot where you want it (like a quiescent state) and turn on the slope function, rotate that slope line to cover the curve, and it reads out the slope of rp in Ohms.
The real cost of a 370B is when repairs are needed.
Time to search for parts; price of parts; having to re-design a circuit due to a lack of a needed part.
To test tubes on a 370B, an amplifier is needed to increase the negative offset voltage, and negative step voltages for testing higher power tubes.
One thing I loved about the 370B: put a dot on a curve, move the dot where you want it (like a quiescent state) and turn on the slope function, rotate that slope line to cover the curve, and it reads out the slope of rp in Ohms.
That’s very kind @smoking-amp , but the list you did about possible issues is hijacking me towards the utracer. I’m worried not to be able to fully recover a unit if something is not working properly. I’ll sleep on it some more nights.I have a spare working CRT here, latest rev. 03 version, but with a small screen/phosphor burn spot at the 0,0 origin, you can have it for free if needed.
Yeah, I haven't seen much on Ebay in the way of parts for a 370B or 371B. Too expensive to buy a second one for parts for sure.The real cost of a 370B is when repairs are needed.
The 576 units have become more abundant than they can sell lately I think, used parts showing up regularly. Still cheaper to just buy another unit for parts, except the same parts may be bad on the parts donors anyway. I haven't had any problems fixing up the units I got, but you have to be a little ingenious about some parts, like the push-buttons. The Tek manual gives detailed parts lists with the manufacturer listed. May not find the exact part currently available, but something close, or search Ebay. I did pick up a spare HV (CRT) power supply off Ebay and checked it. The earlier 576 units had sockets for the TTL ICs and all of the units have sockets for the transistors. Some parts assemblies available:
https://www.ebay.com/itm/145859753650 ( switch assemblies for front panel knobs )
https://www.ebay.com/itm/145859557749 ( this has that worst "Invert" pushbutton switch on it, the long blue one )
https://www.ebay.com/itm/145443118170 ( the HV supply for the CRT )
https://www.ebay.com/itm/145839256530 ( the X-Y amplifiers for the CRT )
https://www.ebay.com/itm/145483073704 ( power supply regulator control board )
https://www.ebay.com/itm/145488175207 ( step generator PC board )
https://www.ebay.com/itm/145527961169 ( the step amplifier, I replaced the power transistors and circuit on it , no point in buying this )
https://www.ebay.com/itm/145725104841 ( CRT )
https://www.ebay.com/itm/364963397308 ( most everything for $100. No bids? )
I haven't had to change out any assemblies on the 576 units I have, other then re-designing the Step Amplifier. Just nice to know you can get parts.
The insides:
Last edited:
I've conversed with the Ebay seller for the last Tek 576 listing above.
https://www.ebay.com/itm/364963397308 ( almost everything for $100. No bids? )
They have 5 of these partly cannibalized units, some have the CRTs, and they all are later models with the IEC cord connector. They are located near Dallas TX. Shipping is the main cost factor. If you live near there, the price is "make an offer". Looks like two selected ones might get you a complete unit. But you'll have to reassemble it. Make sure you get a CRT with mu shield and Bezel, HV power supply and the front door-step test plug-in, among all the other assemblies. If I lived near there I would pick up 2 of them. What a great puzzle to send someone 😵
https://www.ebay.com/itm/364963397308 ( almost everything for $100. No bids? )
They have 5 of these partly cannibalized units, some have the CRTs, and they all are later models with the IEC cord connector. They are located near Dallas TX. Shipping is the main cost factor. If you live near there, the price is "make an offer". Looks like two selected ones might get you a complete unit. But you'll have to reassemble it. Make sure you get a CRT with mu shield and Bezel, HV power supply and the front door-step test plug-in, among all the other assemblies. If I lived near there I would pick up 2 of them. What a great puzzle to send someone 😵
Last edited:
I have a uTracer 3+ and it does have limits however at the cost it's been good value for me.Have you ever tested the utracer? It will for sure be needed to be modded in order to work with positive voltages.
Working within those limits the uTracer 3+ has proven a very capable instrument to accurately collect tube numerical data and graphically display that data.
The pulse method it uses has a big advantage when exploring the "limits" of a tube as the 1.25mS power pulse is unlikely to destroy the sample being tested. The same pulse method can be optimistic with power tubes as the sample does not experience the heating it would in a "real" circuit.
So as in all instruments there are advantages and disadvantageous.
The software Ronald provides I have found easy to use and pretty flexible. There are a few bugs but as is the software way they have for the most part workarounds.
There is also a very long list of modifications and enhancements for the uTracer3+ to extend it ability including a method to test positive grid voltages on triodes.
I would consider a uTracer3+ and if it does not do all you need modify it until it does.
I also have a uTracer6 however I found several accuracy issues with this design currently and so it lost my trust and now mostly collects dust.
I fully understand those that want nothing to do with PC computers and the endless software issues however the uTracer3+ is just so useful...
Hi, I would like to get a confirmation looking at the output transformer performance using different kind of feedback and a pentode as output tube.
To resume:
Roberto
- Local feedback on the output pentode (setting it in triode, UL, Schade, UNSET or others): the rp of the tube is reduced so the output transformer is driven by a lower impedance that implies wider bandwidth, lower distortion, lower Zout of the amp even if the output transformer is outside the nfb loop;
- Global feedback from secondary (classic feedback from secondary to driver): the output pentode drives the output transformer with still a high impedance, but the full amp has a lower Zout, so the bandwidth of the output transfomer in open loop is limited and its distortion is high, but lowered by the nfb;
- Feedback from primary of the output transformer to the driver: the output pentode sees its rp reduced down to few tens of Ohm, so the Zout of the amp is mainly due to the characteristics of the output transformer. Still the output impedance of the driver is not affected by the feedback, so it has to be designed to properly drive the output tube (pentode is then better to be driven).
To resume:
- Every stage that is inside the loop do not see its output impedance reduced except the last stage of the loop.
- Having the output transformer (and all its phase shifts even at low power and core saturation when pushed) out of the nfb loop permits to increase the amount of feedback applied in the loop with lower risks of oscillations.
Roberto
You're not alone, it seems the FUtracer was a frustrated reponse to that issue.I also have a uTracer6 however I found several accuracy issues with this design .
https://futracer.com/about/
Similar to "D-NFB"?I’ve done it on PP and only simulated in SE with what French call “Correction Différentielle”: you take the output signal you want to correct, scale down and subtract the input signal you want to reference it to.
What you get is the difference between the two, that is the distortion (only).
D-NFB Amplifier experiment
- Home
- Amplifiers
- Tubes / Valves
- Strong local feedback + light global or the opposite?