Pioneer RT-707 service & rebuild

[Trevor White]

Thanks again for your input GPS.

The "Low Leakage" requirement was derived from the specs of the original Pioneer quoted part numbers. Mostly I gathered this information from other forums (Tapeheads) where some guys have posted very helpful complete lists of all the capacitors, transistors, some high wattage resistors and zener diodes, and their modern equivalents. I have no knowledge of these things myself, just repeating what others have determined.

Cheers,
Trevor

 

[smurfer77]

Trevor, if those caps are in the signal path, consider films caps instead of electrolytic. Note that polypro is the popular choice but polyester will be the best bet at 10uF due to size. I went film caps anywhere I possibly could - it's better in the signal path, and the caps wont dry out like electrolytics.

 

[Trevor White]

Trevor, if those caps are in the signal path, consider films caps instead of electrolytic. Note that polypro is the popular choice but polyester will be the best bet at 10uF due to size. I went film caps anywhere I possibly could - it's better in the signal path, and the caps wont dry out like electrolytics.

Cheers, thanks for that David.

 

[Trevor White]

Got around to looking at the Servo Amplifier Assembly (RWG-076) board. The parts look something like this:

All of the semiconductor again tested fine so I left those for now. I did replace most of the mylar caps on this board.

Hello again David,

Referring to this post from your re-cap of the Servo Amp board, can you recall which capacitors (type/style) you used to replace the 'Mylar' caps?

Cheers mate,
Trevor

 

[Trevor White]

Got around to looking at the Servo Amplifier Assembly (RWG-076) board.

All of the semiconductor again tested fine so I left those for now. I did replace most of the mylar caps on this board. Before proceeding I did a couple of things. I made a couple of test tone (sine wave) recordings at 1kHz and 3KHz. This board affects the capstan speed, and if i have a recording I will at least have some kind of reference to get back to the speed i had before (a speed reference relative to before the rebuild, which might not have been spot on, but gives me something reasonable to get back to) - thank you GPS for the tip. This board has a bunch of test points. In the manual you will find that the output of the frequency generator (FG), which ultimately determines the motor speed, should be 1360 Hz. I don't have a frequency counter but got ~1370 Hz on my scope.

Hi mate,

I hope you get a chance to see this, but I am trying to duplicate the test you performed above. I own a Siglent SDS1202X-E scope, not dissimilar from your Rigol. But I'll be dammed if I can figure out how to hook the scope up to perform the measurements you made above.

I can see in the manual the references to the measurement points, which are not numbered on the foil side of the board. Did you remove the board to make these tests? Access to the component side whilst the board is mounted looks very difficult.

The measurement points numbered 3 thru 8 appear to be the pins for the molex connector, with points 9, 10 and 11 are three pins extending from the board. So access to the reverse side, the foil side, seems straightforward.

So I was hoping you could give me a tip on just how you hooked up your scope. Where did you place the probes and where did you mount the ground clamps? If I try to touch a probe to test point 8, with my ground clamp on the 707's frame, I get nothing useful at all. In fact I cannot get any waveforms that look even vaguely like those in the manual.

As you can guess I'm a novice at using a scope, so any tips you can offer would be much appreciated. I don't want to bugger anything up. (I had good success with the scope checking my Head Azimuth, but this measurement has me stumped)

Cheers,
Trevor

 

[Trevor White]

Got around to looking at the Servo Amplifier Assembly (RWG-076) board.

The board itself is removed from the metal assembly by taking out four screws from the pcb side of the setup. You will have to unbundle some of the wiring loom to get good access to the board, but you can flip it upside down for easy soldering.

So, I think I can see one reason why I'm having difficulty. If you look at your picture above of the Servo Amp board, you can see that one side (the upper side in the photo) has an indent. This board matched the shape of the foil diagram in the Service Manual. However, on my machine this circuit board is perfectly square, and appears to have its components 'rearranged'? I am trying to perform my measurements with the board in situ, and so far, even using a DMM, I cannot get measurements that seem to match the specs in the manual. It's very confusing.

I did however manage to get my scope hooked up to the TP, Test Point pin. This gave a wave form that vaguely resembled one in the manual, but the voltages were much higher. The frequency reported was 1.36kHz. The manual says approximately 1.36kHz, so I guess that's close .

Can you offer any insight?

Cheers,
Trevor

 

[GPS16]

Hi Trevor,

if the machine is a runner then the only Test Point necessary is TP on the servo PCB. There should be a Pin in the board near to the centre. With the probe tip on the TP Pin and the ground lead on the closest bit of frame there should be a square wave signal of about 5V the square wave itself should be around 5V up from zero or ground. On a bog standard scope you should get a decent trace with the Y set to 1V/Div AC. To check the elevation from ground 5V/Div DC.

Check on which type of probe you have attached. Is it X1 or a X10. As a rule of thumb I prefer to stick to a X10 as this does not load the circuit being measured much. A X1 probe is pretty much the input impedance of the scope (1MOhm/20pF). A X10 probe increases the impedance to10MOhm but then all readings need to be multiplied by 10. Hence the name. If the X10 probe is used then the probe itself needs to be adjusted so that the capacitance of the probe matches the 20pF of the scope. This is done with the calibrator point on the scope. There is usually a 1V pk-pk Square Wave at 1kHz and the probe is adjusted to give a flat top and bottom to the displayed waveform. Your scope may differ slightly from mine which is a HP1740A.

I advised Smurfer to do some recordings from a signal generator before any work was done on the servo PCB so that he could adjust it back to how it was with ease, If you do not have a Speed Test Tape then either a frequency given in the manual will get you close or it is a measured length of tape and a stop watch. This is tedious but still pretty accurate.

Scopes can be a pain to get used to. Initially you will be wasting time searching for the right buttons and modes to get the damn thing to behave and display what you want to see. Again as a rule of thumb there are direct and dual choices which you need to decide. As a sequence I usually go:-
Y-Volts/Div, AC or DC, X-Sec/Div, Trigger Channel, Trigger AC or DC. That should give you some sort of display to get you started.

Check the trigger side of the scope there should be an Auto Mode which should be set to ON. This gives a false trigger to get a sweep with no signal or trigger point and helps to get things in the right ball park as you know that even with a non triggered signal there is a trace somewhere.

Hope this helps.

GPS16

 

[Trevor White]

Hi Trevor,

if the machine is a runner then the only Test Point necessary is TP on the servo PCB. There should be a Pin in the board near to the centre. With the probe tip on the TP Pin and the ground lead on the closest bit of frame there should be a square wave signal of about 5V the square wave itself should be around 5V up from zero or ground. On a bog standard scope you should get a decent trace with the Y set to 1V/Div AC. To check the elevation from ground 5V/Div DC.

Check on which type of probe you have attached. Is it X1 or a X10. As a rule of thumb I prefer to stick to a X10 as this does not load the circuit being measured much. A X1 probe is pretty much the input impedance of the scope (1MOhm/20pF). A X10 probe increases the impedance to10MOhm but then all readings need to be multiplied by 10. Hence the name. If the X10 probe is used then the probe itself needs to be adjusted so that the capacitance of the probe matches the 20pF of the scope. This is done with the calibrator point on the scope. There is usually a 1V pk-pk Square Wave at 1kHz and the probe is adjusted to give a flat top and bottom to the displayed waveform. Your scope may differ slightly from mine which is a HP1740A.

I advised Smurfer to do some recordings from a signal generator before any work was done on the servo PCB so that he could adjust it back to how it was with ease, If you do not have a Speed Test Tape then either a frequency given in the manual will get you close or it is a measured length of tape and a stop watch. This is tedious but still pretty accurate.

Scopes can be a pain to get used to. Initially you will be wasting time searching for the right buttons and modes to get the damn thing to behave and display what you want to see. Again as a rule of thumb there are direct and dual choices which you need to decide. As a sequence I usually go:-
Y-Volts/Div, AC or DC, X-Sec/Div, Trigger Channel, Trigger AC or DC. That should give you some sort of display to get you started.

Check the trigger side of the scope there should be an Auto Mode which should be set to ON. This gives a false trigger to get a sweep with no signal or trigger point and helps to get things in the right ball park as you know that even with a non triggered signal there is a trace somewhere.

Hope this helps.

GPS16

Hi GPS,

Many, many thanks for taking the time to explain this. I am using my probes in X10 mode, and have calibrated them using the scopes inbuilt 1kHz generator. All good there.

I had in fact found that TP pin and hooked up to it. The readings I got from the scope though were way higher than I expected. I have now learned that when setting the Y parms in the scope I should use a probe setting of X1, on the scope itself, the probes themselves are still X10. I had previously set it to X10, thinking that was supposed to match the setting on the probe. However this seems to be some kind of 'multiplier' and all the measurements were multiplied by 10!! Now that I have set this parameter back to X1 I get a peak to peak voltage of 10.16V and a frequency of 1.36kHz (which is spot on). The waveform very closely matches figure 37 in the manual, although I did find I needed to use 2V/Div AC.

Hopefully this means all is good. I am a bit surprised however that the Servo board in my machine seems physically different from the pictures taken by Smurfer.

Thanks again,
Trevor

 

[GPS16]

Hi Trevor,

there are variations on the theme here. 2 versions of Servo are in the field. There is one where the Servo PCB has its own regulated supply and a later one where a simple Zener Diode and resistor is used. as the Servo loading never changes (by much), the later version is sufficient. I have a circuit for the differences but not a layout. If you search through other 707 threads the information is in one of them. (707 info and tips). You will find one of the pots is missing as there is no regulated voltage to tweak.

If you have a running servo then at this moment in time this issue will just be data overload. Grab the info by all means and file it.

Cheers.

GPS16

 

[Trevor White]

Thanks again GPS. Wise words re the data overload.

My version of the Servo board seems to have 3 trim pots two near the top edge of the board, and another set just behind them. As you say, my servo seems to be operating correctly, so I will just carefully replace the electrolytics and leave well enough alone.

Cheers.

 

[Trevor White]

Playback Amplifier Assembly

However, my list above was slightly different to what was inside the machine. C129 is a ceramic listed at 0.01 uF, 50 V, but inside my machine I found a 0.047uF. Also the two VRs for the VU meters that are mounted onto this board (VR109/110) are listed as 10 kOhm but inside my machine i found 6.8 kOhm devices. Small stuff, but there ya go. They 20kOhm VRs grayed out in the list are rear volume controls for the line-out , which I will not be replacing. The other VRs functions are outlined in the pic I made in an earlier post.

The new VRs are Bourns multiturn devices - it's a dream compared to my old pots which were really sticky and single turn. When replacing the VRs, I made some measurements of the settings/values of the original VRs and set the new ones to approximately the same values. My notepage has a page of weird looking sketches that correspond to the VRs and their placement in the circuit so I know which way up the drawing is... if anyone wants the values or pics, let me know. It saved me a lot of time on the playback setup by having the new VRs set at reasonable values.

Hi there Smurfer77,

I was contemplating also getting myself some of those lovely little Bourns VR's. I was wondering if you still had those diagrams you made? If so they may be of assistance to me.

Also, if you still have the details of the Mouser part numbers for those multi-turn devices, that would be very helpful also. Looking on Mouser it is tricky to choose. The 3296Y series don't seem to have a 33Kohm version which is listed on the Circuit diagrams. What did you actually use?

Cheers,
Trevor

 

[smurfer77]

Hi there Smurfer77,

I was contemplating also getting myself some of those lovely little Bourns VR's. I was wondering if you still had those diagrams you made? If so they may be of assistance to me.

Also, if you still have the details of the Mouser part numbers for those multi-turn devices, that would be very helpful also. Looking on Mouser it is tricky to choose. The 3296Y series don't seem to have a 33Kohm version which is listed on the Circuit diagrams. What did you actually use?

Cheers,
Trevor

Trevor, it looks like I didn't scan my notebook yet, I shall do so, but the VR diagrams are going to be hard to make sense of for anyone but me. What I recommend is, before you remove your VRs, just make some quick measurements (before dismounting) and familiarize yourself and make some notes. I don't recall for now which were voltage dividers (pay attention to ratio of 'arms') and which were single VR scenarios (with one 'arm' shorted) so pay attention with some measurements. If you get lost and I haven't posted the sketches, just hassle me again and I will speed up . Note that the PCB is a bit different to the circuit diagram and this can be confusing. For example, some of the VRs are shown with input coming into one end of VR, and then a trace shorting out between this input and the middle wiper; on the PCB the input actually just goes directly to the wiper skipping the first arm of the VR - same thing but confusing if you try to follow the traces with the circuit diagram at hand. Also note that measurements in circuit and out-of-circuit can be different due to nearby parallel resistors, shorting, etc.

The other point is an easier one: I used 50k VRs in place of the 33k Ohm. I also noted that the two 10k Ohm VRs were actually/originally 6.8kOhm on my PCB (I ordered and used 10kOhm). Just make sure you get bigger value, not smaller value. Yes I used 3296Y series. The multiturn made adjustments so much easier and stable.

22kOhm: 3296Y-1-253LF (EQ/gain)
50kOhm: 3296Y-1-503LF (EQ/gain)
10kOhm: 3296Y-1-103LF (VU meter)

 

[Trevor White]

Trevor, it looks like I didn't scan my notebook yet, I shall do so, but the VR diagrams are going to be hard to make sense of for anyone but me. What I recommend is, before you remove your VRs, just make some quick measurements and familiarize yourself and make some notes. I don't recall for now which were voltage dividers (pay attention to ratio of 'arms') and which were single VR scenarios (with one 'arm' shorted) so pay attention with some measurements. If you get lost and I haven't posted the sketches, just hassle me again and I will speed up . Note that the PCB is a bit different to the circuit diagram and this can be confusing. For example, some of the VRs are shown with input coming into one end of VR, and then a trace shorting out between this input and the middle wiper; on the PCB the input actually just goes directly to the wiper skipping the first arm of the VR - same thing but confusing if you try to follow the traces with the circuit diagram at hand. Also note that measurements in circuit and out-of-circuit can be different due to nearby parallel resistors.

The other point is an easier one: I used 50k VRs in place of the 33k Ohm. I also noted that the two 10k Ohm VRs were actually 6.8kOhm (I ordered and used 10kOhm). Just make sure you get bigger value, not smaller value. Yes I used 3296Y series. The multiturn made adjustments so much easier and stable.

22kOhm: 3296Y-1-253LF (EQ/gain)
50kOhm: 3296Y-1-503LF (EQ/gain)
10kOhm: 3296Y-1-103LF (VU meter)

Fantastic. I really appreciate that, thanks.

EDIT: I have just had a close look at the diagrams. I see what you mean! The Foil side diagram shows the Wiper and one end of the VR linked on all of the 22K EQ trimmers.

 

[Trevor White]

Hello again David,

Well, thanks to your help, I have obtained my Bourns multi-turn VR's. The same versions you provided part numbers for above.

I have spent the afternoon carefully making my own diagram of the way that the original VR's were mounted and how they were connected. My circuit diagram (foil side) actually matched how the VR's were mounted. I took measurements of the resistance values between the Wipers and the other pins. On the VR105 - VR108 set one of the pins (pin 3) was actually connected to ground on all of the VR's. The resistance values on all of these between the Wiper (pin 2) and Ground (pin 3) was 29.6K, 30.1K, 22.7K and 25.5K. The values between the Wiper (pin 2) and pin 1 was the remainder of the total resistance value of the VR, which for the 33K VR's were values around 4.8K to 6.6K depending on the overall resistance of the VR's. Makes sense I guess.

Each of those Pin 1 connections was connected to Pin 3 of the EQ VR's VR101, VR102 and VR103, for VR104 Pin 1 of VR108 was actually connected to Pin 1 of VR104.

For each of the EQ VR's, VR101 - VR104 the Wipers (pin 2) and pin 3 was connected (separately) to pins on the Head Amp assemblies. In each case with VR101 - VR104 the Wiper (pin 2) was interconnected with pin 1.

I measured each of the resistances between the Wiper and both the other pins. The resistance values on all of these between the Wiper (pin 2) and pin 3 was 6.72K, 7.55K, 6.74K and 6.54K. The values between the Wiper (pin 2) and pin 1 was the remainder of the total resistance value of the VR which for these 22K VR's were values around 12.3K to 17.4K depending on the overall resistance of the VR's.

The problem I am having is just figuring out which resistance value I need to use when initially setting up the Bourns VR's. If I use the larger value measured from the old VR's then the remainder value on the other pin will be a lot higher as the Bourns VR's have a much higher overall resistance. Whilst if I set the 'remainder' value to match what it was on the old VR's, then the other value will be correspondingly higher.

I'm not that familiar with using these kinds of devices and despite studying the diagrams I'm just not sure which is the critical value I need to use to provide the initial setting of the Bourns VR's.

I don't know if any of this makes any sense, it's hard to describe without a diagram, but can you recall what you did with your machine? Or can you provide some clarification, despite my convoluted description.

Cheers mate,
Trevor

 

[GPS16]

Hi Trevor,

If the VR has the wiper shorted to one end, then the resistance between the wiper and the end that is free is what is important. If the replacement is a different value then make sure the "difference" is in the section shorted by the wiper to the end.

If both ends go to the circuit then you need to work out the ratio between the two resistances and then translate that to the new resistance if different overall value.

If you intend to do a proper realignment then you will be setting them anyway. This will get you close to what it was. After the realignment it will be spot on!!! That is the point of the Test Tape and the Millivoltmeter, Sig Generators...........

It will get the "familiarity with the beast" up a couple of notches though!!!

GPS16

 

[calman46]

So the new trimpots have a larger value so you can't get both sides to the match orignal trimpot measures. I think you're just going to have to tweak it to get as close as you can to what measures you want. You can lock it in when you calibrate the deck . The original values are not available ?

 

[Trevor White]

Hi Trevor,

If the VR has the wiper shorted to one end, then the resistance between the wiper and the end that is free is what is important. If the replacement is a different value then make sure the "difference" is in the section shorted by the wiper to the end.

If both ends go to the circuit then you need to work out the ratio between the two resistances and then translate that to the new resistance if different overall value.

If you intend to do a proper realignment then you will be setting them anyway. This will get you close to what it was. After the realignment it will be spot on!!! That is the point of the Test Tape and the Millivoltmeter, Sig Generators...........

It will get the "familiarity with the beast" up a couple of notches though!!!

GPS16

Excellent GPS, many thanks. That clears it up nicely.

I was wondering about your earlier comments about the 'Ratio of Arms' which initially I did not understand. But it becomes clearer now. I took careful measurements of all the VR's both in the circuit and out of it, so I should be able to get close.

Cheers,
Trevor

EDIT: Ooops, I just realised that it was not Smurfer that provided the advice above, but GPS16. Sorry mate, many thanks for your input.

 

[Trevor White]

So the new trimpots have a larger value so you can't get both sides to the match orignal trimpot measures. I think you're just going to have to tweak it to get as close as you can to what measures you want. You can lock it in when you calibrate the deck . The original values are not available ?

Hi mate, unfortunately the 33K VR's specified for the circuit are not available in the Bourns series. So I have 50K, same as used by Smurfer77. So it is just going to take a little calculation to get close. And as GPS16 says it will all be re-calibrated anyway.

Cheers.

 

[GPS16]

Hi Trevor,

no problem.

I had sussed what had happened!!! This thread is Smurfer's baby after all.

GPS16

 

[smurfer77]

Thanks GPS for taking the questions here .

Yes, Trev, in my incredibly brief sentence, "don't recall for now which were voltage dividers (pay attention to ratio of 'arms') and which were single VR scenarios (with one 'arm' shorted) so pay attention with some measurements", I did mean as GPS clarified in more detail in post #195

And I see you have no uploaded a sketch of the VR rough setup before uninstalling. For those unaware of that thread check it here. I'm also going to edit my first post of this thread, with a link to your thread.